OR/14/004 Worst case scenario research at FMI

From Earthwise
Jump to navigation Jump to search
Thomson, A W P (Editor), Beggan, C, Kelly, G, Baillie, O, Viljanen, A, and Ngwira, C. 2014. Project EURISGIC: worst case scenarios (Technical note D5.1). British Geological Survey Open Report, OR/14/004.

GIC in the Norwegian high-voltage power grid[edit]

As an exploitation of existing EURISGIC results, FMI performed an assessment of GIC in the Norwegian high-voltage power grid, together with the Statnett company in 2012–2013 (Myllys et al., 2013[1]). The basic results were statistics of the geoelectric field and of GIC based on 10-s geomagnetic data measured in 1994–2011 and on ground conductivity models derived by the EURISGIC project team.

The statistics are based on an 18-year period, which corresponds to about 1.5 sunspot cycles. This is a relatively short time and it does not contain the largest known geomagnetic storms. However to extend the results, we extrapolated the 18-year distributions to determine a once in 100-year case.

We calculated the 10-s electric field values at selected magnetometer stations using the data from 1994–2011. We fitted two curves to the electric field values to estimate the largest magnitude of the 10-s value occurring once in 100 years. Figure 3.1 shows an example for the Tromsø observatory. As a rule of thumb, the once in 100 years field is expected to be about 1.5–2 times larger than that modelled in the period of 1994–2011. The same holds for modelled values calculated from 1-min averages of the magnetic field.

Figure 3.1    Statistical occurrence of 10-s electric field values in 100 years at the Tromsø observatory. Coloured lines correspond to two different extrapolation methods.

Table 3 shows the modelled electric field values at five magnetic observatories during three major storms before 1994. We used 1-min magnetic field values and an identical 2-layer ground conductivity model at all locations. The upper layer of the ground model is 200 km thick and its resistivity is 5000 ohm m. The resistivity of the lower layer is 200 ohm m. With this ground model, the estimated once in 100 years electric field values at Norwegian magnetometer stations vary between 2 and 10 V/km when determined from 1-min magnetometer data. The largest 1-min field value in 1994–2011 with the same ground model is 4.5 V/km in North Norway.

As Table 3 shows, the electric fields on 13–14 July 1982 reach values equal to the estimated once in 100 year event in Norway at a wide latitude range from Denmark to North Finland. There is also a notably large localized peak value at Brorfelde during the March 1989 storm. Maximum values during the March 1991 storm are also comparable to the Norwegian maxima in 1994–2011.

It seems that the 18-year period of 1994–2011 is too short for assessing extreme magnetic storms. In particular there are at last three magnetic storms in 1982–1991 during which the modelled electric field reaches or exceeds the modelled maximum values in 1994–2011.

Table 3    Maximum time derivative of the horizontal magnetic field vector and maximum of the modelled horizontal electric field at five observatories during three magnetic storms. In all cases, 1-min magnetic field data were used. The same ground conductivity model was assumed at all sites (see text for details).
13–14 July 1982
Observatory Max(|dH/dt|) [nT/s] Max(Ehor) [V/km]
Sodankylä (FI) 33.3 7.78
Nurmijärvi (FI) 19.9 8.04
Lovö (SE) 44.8 8.67
Brorfelde (DK) 24.5 6.79
Wingst (DE) 8.0 2.52
13–14 March 1989
Sodankylä (FI) 10.4 3.71
Nurmijärvi (FI) 10.8 3.50
Lovö (SE) 11.7 3.71
Brorfelde (DK) 33.2 9.00
Wingst (DE) 10.7 2.84
24–25 March 1991
Sodankylä (FI) 21.0 4.58
Nurmijärvi (FI) 15.7 5.17
Lovö (SE) 14.1 4.61
Brorfelde (DK) 7.6 1.89
Wingst (DE) 6.2 1.29

Results from Russian geomagnetic recordings in 1850–1862[edit]

We have analysed geomagnetic recordings (Viljanen et al, 2013[2]) at four subauroral and midlatitude Russian observatories in 1850–1862 (Figure 3.2). The data consist of spot readings made once per hour of the north and east components of the magnetic field. We use the hourly change of the horizontal field vector as the measure of activity. We compare these values to data from modern observatories at corresponding magnetic latitudes (Nurmijärvi, Finland; Tartu, Estonia; Dourbes, Belgium) by reducing their data to the same 1-hour sampled format.

Figure 3.2    Russian geomagnetic observatories in the 1800s. Circled stations were used in this study. Note that there are no digital data from PEK (Peking): there are yearbooks of 1851–1855 but with incomplete coverage.

Geographic and geomagnetic coordinates (CGM in 2000 and 1900) for the circled sites in Figure 3.2 are:

Code full_name lat lon latm2000 lonm2000 latm1900 lonm1900

STP St. Petersburg 59.93 30.30 56.16 106.81 54.73 109.17
EKA Ekaterinburg 56.82 60.58 52.72 133.76 50.39 131.93
BAR Barnaul 53.33 83.95 48.97 156.53 47.22 152.74
NER Nertchinsk 51.32 119.60 45.87 192.42 45.30 188.02

The largest variations at the Russian observatories occurred during the Carrington storm in September 1859 and they reached about 1000 nT/h, which was the instrumental off-scale limit (n.b. identical instruments were used at all the sites). When the time stamp for the spot readings happens to be optimal, the top variation in the Nurmijärvi data is about 3700 nT/h (July 1982), and at Tartu the maximum is about 1600 nT/h (November 2004). At the mid latitude site Nertchinsk in Russia (NER in Figure 3.2; magnetic latitude ~45 N), the variation during the Carrington storm was at the off-scale limit (Figure 3.3), and exceeded the value observed at Dourbes (magnetic latitude ~46 N) during the Halloween storm in October 2003. At Nertchinsk, the Carrington event was at least four times larger than any other storm in 1850–1862. The maximum dB/dt at Dourbes during the Halloween storm was about 900 nT/h, so it was smaller than at NER during Carrington.

Despite the limitations of the old recordings and in using only hourly spot readings, the Carrington storm was definitely a very large event at midlatitudes. At higher latitudes, it remains somewhat unclear whether it exceeds the largest modern storms, especially the one in July 1982.

Figure 3.3    One-hour values of the time derivative of the horizontal magnetic field vector at Nertchinsk in 1851–1862. The largest value of at least 1000 nT/h occurred during the Carrington storm in 1859.


  1. MYLLYS, M, VILJANEN, A, ARUI, Ø A, and OHNSTAD, T M: Geomagnetically induced currents in Norway: the northernmost high-voltage power grid in the world. Submitted to Journal of Space Weather and Space Climate, 2013.
  2. VILJANEN, A, MYLLYS, M, and NEVANLINNA, H. 2013. Russian geomagnetic recordings in 1850–1862 compared to modern observations. Submitted to Journal of Space Weather and Space Climate.