Chronology of effects
Many different man-made systems are affected during magnetic disturbances. Here we have only attempted to document the effects of geomagnetic disturbances on electrical systems at the Earth's surface, such as power systems, pipelines, and communication cables. Our list only shows incidents that can be related to a specific magnetic disturbance, and it is likely that some failures, due to cumulative effects, have never been related to geomagnetic influence.
Barlow (1848) was the first person to make systematic observations of the "Spontaneous Electrical Currents observed in the Wires of the Electric Telegraph":
When the telegraph instruments are not working, the batteries are put out of circuit, and the wires remain with a simple earth connection at both extremities.
It was in this condition of the wires that spontaneous currents were observed to arise in them, producing occasionally large deflections in the needles. These deflections were sometimes to the right and sometimes to the left; at times they changed rapidly from right to left, at others they continued in one direction from periods varying from a few minutes to one or more hours.
My attention was strongly drawn to the subject by the constancy of these effects, when a circumstance occurred which imparted a new interest to the inquiry. On the evening of the 19th of March, 1847, a brilliant aurora was seen, and during the whole time of its remaining visible, strong alternating deflections occurred on all the instruments. Similar effects were observed also on the telegraphs on several other lines of railway.
October 27-28 was the occasion of heavy disturbances on the telegraph (Burbank, 1905). Matteucci remarked on the coincidence of the aurora with interruptions in telegraphic communications, produced by telluric currents.
November 17. Appearance of the aurora coincided with effects on the electric telegraph between Florence and Pisa (Prescott, 1866, p 317).
Prescott (1866) reports,
a remarkable aurora, which took complete possession of all the telegraph lines in New England, and prevented any business from being transacted during its continuance.
On February 19 a brilliant auroral display was observed. Associated with it were currents that burnt through the chemical paper used with the Bain's chemical telegraph in the northeastern US (Prescott, 1866, p 318).
Varley (1873) describes some of the early earth current observations:
In 1853 or 1854, the earth current on a length of two miles of wire between Telegraph Street and the Strand was so strong that is was impossible to work the line during its continuance.
A magnetic storm from August 28 to September 2 produced widespread effects on the telegraph system in Europe and North America. Mr O. S. Wood, Superintendent of the Canadian telegraph lines, says (see Prescott, 1866):
I never, in my experience of fifteen years in the working of telegraph lines, witnessed anything like the extraordinary effect of the aurora borealis, between Quebec and Father's Point, last night. The line was in most perfect order, and well-skilled operators worked incessantly from eight o'clock last evening till one o'clock this morning, to get over, in even a tolerably intelligible form, about four hundred words of the steamer Indian's report for the press; but at the latter hour, so completely were the wires under the influence of the aurora borealis, that it was found utterly impossible to communicate between the telegraph stations, and the line was closed for the night.
Problems were also reported by telegraph operators in New York, Washington, Philadelphia, Vermont and Massachussetts (Prescott, 1860), 1866; also see Clement, 1860). Interesting observations were made in France by Bergon and Blavier. Blavier's notes, reproduced by Angot (1897) report that:
At all the telegraphic stations in France the service was impeded during the whole of September 2,  but especially at two periods of the day, from 4:30 A.M. to 9 A.M., and from noon to 3 P.M. These two periods were the same at all stations, and the greatest disturbances took place exactly at the same hours, at 7 A.M. and at 2 P.M. The longest wires always showed the greatest disturbances. The same day telluric currents were also observed in the greater part of the two hemispheres, in Switzerland, in Germany, in the British Isles, in North America, and throughout Australia.
Many reports from telegraph operators were collected by Silliman and published in the American Journal of Science and Arts, 1860.
Observations made at Boston, Mass., and its vicinity by George B. Prescott, Telegraph Superintendent.
The effects of the magnetic storm of August 28th, 1859, were apparent upon the wires during a considerable portion of Saturday evening, and during the entire day, Sunday. At 6 P.M. the line to New Bedford (60 miles in length, running a little west of south) could be worked only at intervals, although, of course, no signs of the Aurora Borealis were visible to the eye at that hour. The same was true of the wires running east through the state of Maine as well as those running north to Montreal.
On Friday, September 2d, 1859, upon commencing business at 8 o'clock A.M. it was found that all the wires running out of the office were so strongly affected by the auroral current as to prevent any business being done, except with great difficulty.
Observations made at Philadelphia, communicated by H. Emmons Thayer, Telegraph Superintendent.
On the evening of Aug. 28th, about 8 o'clock, we lost current on all our four wires running from Philadelphia to New York, and we had strong circuit as if from a near ground connection; but there was no interruption on wires running south to Baltimore and Washington. At 9h10m the wires were relieved to a great extent from the influence of the Aurora, giving us our usual working current.
On testing wires at 8 o'clock on the morning of Sept. 2d, I found two of our wires, those running via Camden and Amboy to New York, strongly under the influence of an Aurora. The effect was different from that of Aug. 28th. There was an intensity of current which gave a severe shock when testing, giving a reversed current, neutralizing our batteries, and destroying magnetism. On removing the batteries, we had a very strong circuit, giving powerful magnetism, but could not raise New York. On the line running from this city to Pittsburgh, the operator, Mr. Steacy, succeeded in transmitting a business message to Pittsburgh wholly on the auroral current. The current was changeable, suddenly disappearing and returning at intervals of from five to ten minutes. The signals were distinct and the conversion lasted four or five minutes, the operators exchanging remarks as to the singularity of the phenomenon. At 9. A.M. all the wires were relieved from the effects of the Aurora, and worked as well as usual.
Observations made at Washington, D.C., by Frederick W. Royce, Telegraph operator.
On the evening of Aug. 28th I had great difficulty in working the line to Richmond, Va. It seemed as if there was a storm at Richmond. I therefore abandoned that wire, and tried to work the northern wire, but met with the same difficulty. For five or ten minutes I would have no trouble, then the current would change, and become so weak that it could hardly be felt. It would then gradually change to a 'ground' so strong that I could not lift the magnet. The Aurora disappeared at a little after 10 o'clock, after which we had no difficulty. During the auroral display, I was calling Richmond, and had one hand on the iron plate. Happening to lean towards the sounder, which is against the wall, my forehead grazed a ground wire. Immediately I received a very severe electric shock, which stunned me for an instant. An old man who was sitting facing me, and but a few feet distant, said that he saw a spark of fire jump from my forehead to the sounder.
Observations made at Pittsburgh, Pa., communicated by E.W. Culgan, Telegraph manager.
During the Aurora of Aug. 28th the intensity of the current evolved from it varied very much, being at times no stronger than an ordinary battery, and then suddenly changing the poles of the magnets it would sweep through them, charging them to their utmost capacity, and compelling a cessation of work while it continued.
On the morning of Sept.2d, at my request the Philadelphia operator detached his battery, mine being already off. We then worked with each other at intervals as long as the auroral current continued, which varied from thirty to ninety seconds. During these working intervals we exchanged messages with much satisfaction, and we worked more steadily when the batteries were off than when they were attached.
On the night of Aug. 28th the batteries were attached, and on breaking the circuit there were seen not only sparks (that do not appear in the normal condition of a working line) but at intervals regular streams of fire, which, had they been permitted to last more than an instant, would certainly have fused the platinum points of the key, and the helices became so hot that the hand could not be kept on them. These effects could not have produced by the batteries.
Observations at Christiania, Norway, by Prof. Christoph Hansteen
The effect of this aurora upon the telegraph lines in Norway was much greater than in France and Germany. The effect was noticed from the opening of the stations at 7 A.M. On the 29th communication was interrupted till 11 A.M. on almost all the lines; and likewise Sept. 2d, but with a long repetition after 2 P.M. Sept. 3d, only towards 8-1/2 A.M.
Effect on the telegraph Wires, from the Comptes Rendus, T. XLIX, p. 365
From the evening of Aug. 28th until the morning of the 29th the needles of the magnetic telegraph at Paris were almost constantly in motion, as if a permanent current was passing through the telegraph wires. Business was therefore entirely interrupted, and could not be resumed until 11 A.M. Aug. 29th. The same effect was noticed on the telegraph lines from 4h to 8h on the morning of Sept. 2d, although no aurora was noticed on that day. Business was again interrupted, the needles were disturbed, and the bells were rung.
Observations at Brussels by M. Quetelet
About midnight Aug. 28th, the employees in the telegraph office at Brussels noticed signals from their bells, such as often occur during a storm. The employees in the offices at Mons, Antwerp, Gand and Ostend were also awakened by their bells, and enquired what was wanted. Communication with Paris, London, and Berlin were interrupted till 1h 30m. Paris and London inquired of our operators if they saw a light in the heavens. The effect ceased at 1h 30m on all the lines except the submarine line from Ostend to Dover, which was charged with electricity throughout the entire morning. It was not till 3h 30m, and after nearly doubling the battery, that communication was re-established.
September 2, between 5h and 6h A.M., there was a second disturbance on all the telegraph lines, and communication between Brussels, Paris, and London was interrupted.
On May 30, 1869, during the aurora borealis which was visible in Switzerland from 7 to 9 p.m., of the sixteen lines which terminated in the telegraphic office at Basle, six were almost useless during the two hours that the phenomenon lasted; on the others the telluric currents were not strong enough absolutely to interrupt communication (Angot, 1897, p 141).
Disturbances to telegraphic communication reported by Angot (1897):
The telluric currents attained an extraordinary development during the aurora of February 4, 1872, which we have mentioned as one of the most extensive known; it was seen in the whole of the West of Asia, in the North of Africa, throughout Europe, and on the Atlantic as far as Florida and Greenland; at the same time an aurora was observed in part of the southern hemisphere. The disturbances in telegraphic communication were not less extensive, and were observed with great care, in great part of Europe. At the same time many of the submarine cables were so affected as to prevent the transmission of any messages; the disturbance was especially marked on the line from Lisbon to Gibraltar, on the Mediterranean cable, on the line from Suez to Aden, and from Aden to Bombay, and finally along the Transatlantic cable from Brest to Duxbury.
Burbank (1905) reports that on January 31 the telegraph in England experienced a current corresponding to a potential difference of 1.1 volts per kilometer.
Telluric currents observed in England on November 17 were, according to W.H. Preece [Superintendent of the Telegraph] five times as strong as the current usually employed in telegraphy. Communication was interrupted as long as the disturbance lasted (Angot, 1897).
Finn (1903) reports,
During an electrical disturbance in 1891 similar to that through which we have just passed an electromotive force of 768 volts was recorded on the Western Union lines between New York and Buffalo, the circuits varying from 450 to 480 miles in length. On several occasions the strength of the earth current reached nearly 300 milliamperes. The average strength of the working currents does not exceed 35 milliamperes.
On July 16 there was a serious interruption of wire service in US (Sanders, 1961).
Telegraph operators had been supplied with telephones and heard a variety of sounds produced by earth currents in the lines. Preece (1894) recounts:
Peculiar and weird sounds distinctly perceived, some highly-pitched musical notes, others resembling murmur of waves on a distant beach.... The musical sounds would very much resemble those emitted by a number of sirens driven at first slowly, then increased until a 'screech' is produced, then again dying away. Duration of each averaged about twenty seconds.
Lockyer (1903) reports that on October 31 practically the world's whole telegraph system was upset:
The enormous development of the telegraph, telephone, cable, and other applications of electricity since the date of the last great magnetic storm has caused the disturbance to be more generally observed that was previously perhaps the case.
Practically the worlds whole telegraph system was upset, and information from this country, France, and the United States and other lands shows that for several hours communication was almost completely interrupted.
On September 25, telegraph disturbances occurred in many parts of the world and were accompanied by auroral displays down to magnetic latitudes of 30 (Silverman, 1995). In Western Australia, the telegraph lines between Perth and Kalgoorlie, a distance of 350 miles, and between Perth and Albany worked with all the batteries cut out for half an hour on the evening of 25 September. A magnificent visual aurora was also present, and the telegraph currents waxed and waned concurrently with the auroral light (Nature, 81, 524, 1909).
The geomagnetic storm caused problems in the Swedish telegraph system. Lundstedt (2006) quotes the Ph.D. thesis of D. Stenquist, published in 1914, which thoroughly documents the storm:
Strong earth-currents disturbed the telegraphic communications over the whole globe on September 25th… Earth-currents were first observed at the Telephone office on the wires to Luleå. The operator on duty at these wires received a sudden shock, half her hand turning white and two of her fingers being paralyzed. When taking hold of a microphone, both the instrument and the hand were surrounded with an intense, diffuse light, casting out sparks and causing blisters.
No reports found.
Karsberg et al (1959) report that induced currents caused fires in telegraph equipment in Sweden. Exact date is not given, but Chapman and Bartels (1940) show that a great magnetic storm occurred on May 13 - 15, 1921.
On April 16 there were problems on telegraph system in Norway (Harang, 1941).
The magnetic storm and its effects are described in a group of papers by McNish (1940), Germaine (1940) and Davidson (1940). The effects on communications are recounted by Germaine (1940):
When the magnetic storm of Sunday, March 24, 1940, developed, it immediately became apparent that it was unusual intensity and covered a wide area, interrupting nearly all overseas radiotelephone circuits, service to ships at sea, and a number of long distance land telephone and other communication services.
While most of the serious effects on the services on the land lines took place between 9:00 a.m. and 3:00 p.m. Central Standard Time, the earth current disturbances persisted until the morning of the 25th and disturbances continued to intermittently affect overseas service for a number of days, with the Transatlantic service the most seriously affected.
The most seriously affected sections of the country were on the Eastern Seaboard and in the states of North Dakota, Minnesota, Iowa, and Missouri. Less serous effects were felt as far south as Georgia and as far west as Arizona, Colorado, and Montana. This is shown rather clearly in Figure 1.
During the storm, failures to less than 10 per cent of the long distance telephone land circuits throughout the country were experienced; however, out of Minneapolis about 80 per cent of the long distance circuits were out of service at one time during the day; while out of Chicago, more than 200 circuits could not be used between noon and about 2 p.m.
This storm also marks the first time that effects on power systems were reported. These are summarised by Davidson (1940):
The basic information was obtained from 22 power companies scattered about the United States and in Canada. Disturbances on power systems were reported by companies in New England, New York, eastern Pennsylvania and southern and eastern parts of Minnesota and in Ontario and Quebec. Reports of "No trouble" were received from systems in North Carolina, Georgia, Alabama, Illinois, Nebraska, Kansas, northern Minnesota, Utah, Idaho, Washington and California.
Of the 10 systems on which disturbances were noted, the effects, listed in the inverse order of the number of cases reported, were:
- Voltage dips ranging up to 10 per cent but generally of short duration (7 cases).
- Tripping transformer banks by differential relay operation (5 cases involving 15 transformer banks).
- Large increases or swings in reactive kva (4 cases).
High neutral currents and tripping of transformers occurred in Ontario (Acres, 1975).
On March 28, 1946 between 0646 and 0717 hours EST, the Port Arthur transformer tripped six times and the Crow River transformer twice; it was reported that the telephone relays chattered quite vigorously just preceding each operation of the transformer differential relay.
On September 22, 1946 a maximum current of 93 amp was measured in a transformer neutral at Port Arthur. The Port Arthur transformer differential relay protection operated eight times between 0515 and 0946 hours EST.
On January 21, disturbances occurred on power feeding circuits on transatlantic submarine cables on (Axe, 1968).
On September 22, a power system 230kV breaker tripped due to saturation of transformer cores and excessive 3rd harmonic currents in ground relays. (Slothower and Albertson, 1967)
On February 10, a magnetic storm caused disruption of service on the TAT-1 transatlantic cable (Winckler et al, 1959). Abnormal power flows occurred in the power system in Minnesota (Slothower and Albertson, 1967). Lanzerotti and Gregori (1986) report that the Toronto area suffered from a power blackout. However Acres (1975) report that:
On February 10, 1958 at exactly 2104 EST, a transformer at Rayner Generating Station and the Port Arthur transformers T1 and T2 were simultaneously tripped at Port Arthur, loss of load occurred for about 4 minutes. However, Ontario Hydro does not know of any significant loss of revenue or any problems to the customers served by this station.
On November 13, disturbances occurred on power feeding circuits on Transatlantic cables (Axe, 1968). The power system in Sweden experienced tripping of 30 line circuit breakers (Tillberg and Andersson, 1977; Elovaara et al, 1992).
A major magnetic storm occurred on August 4 and caused an outage of the L4 cable system in the continental US (Anderson et al, 1974) and problems on power systems (Albertson and Thorson, 1974). An investigation of the L4 outage by Anderson et al (1974) found that a rapid change of the magnetic field coincided with the sytem shutdown. They calculated that this produced an induced electric field along the cable of approximately 7.0 V/km which exceeded the ~6.5 V/km threshold at which the line would experience a high current shutdown. At the time of the outage, satellite observations showed that there was a severe compression of the Earth's magnetic field due to an enhanced flow of plasma in the solar wind. For a long time it was thought that electric currents on the magnetopause, caused by this enhaced solar wind, were responsible for the magnetic disturbance and the L4 outage. However, recent work by Boteler and Jansen Van Beek (1999) has shown that the disturbance was too localised to have been caused by magnetopause currents and is instead consistent with a rapid enhacement of electric currents in the iononosphere.
The effects on power systems in Canada were summarised by Acres (1975):
- Newfoundland and Labrador Power Commission, St. John's, Newfoundland
- The Corner Brook transformer tripped on differential relay (IAC, no harmonic restraint) and was returned to service several times. Long Harbour Terminal tripped as above twice.
- Hydro-Québec, Montréal, Québec
- MW variations of a few per cent. Bersimis No. 2, on power-frequency control, had MW variation of 90 per cent. Significant variations of Mvars at all generating stations. Voltage variations of 3.6 per cent (735 kv sending end) to 5.7 per cent (315 kv receiving end). Shunt capacitor bank tripped off by overload protection at 1852 hours EDT on August 4.
- Ontario Hydro, Toronto, Ontario
- Drop in power on the Ontario - Michigan interconnection from 540 to 480 Mw, and on the Ontario - New York State interconnection from 310 to 280 Mw. Voltage variations of up to 8 per cent. and a change in system frequency from 60.00 to 60.03 hertz.
- Manitoba Hydro, Winnipeg, Manitoba
- Fluctuations on the U.S.A. Tie Line (La Verendrye) from 120 to 164 to 44 Mw and +25 to -100 Mvar. Drop in voltage at Grand Rapids, La Verendrye, Morden, Parkdale and Selkirk.
- At la Verendrye, SIC > 100 amp in neutral of 230/115-kv transformer and at Grand Rapids, SIC ~ 100 amp in transformer. System frequency varies from 60.00 to 59.95 to 60.08 hertz.
- Seven Sisters Unit No. 2 (25 Mw) tripped by generator field ground and overcurrent time A and C phase relays. Slave Falls Unit No.3 (9 Mw) tripped by undervoltage relay. Load Dispatch requested all generator units to maximum voltage boost at 1755 hours CDT, hoping to prevent loss of additional generator unit.
October: trip of 500 kV line from Manitoba to Minnesota (Aspnes et al, 1981).
April: trip of 500 kV line (again) from Manitoba to Minnesota (Aspnes et al, 1981).
July 13: Four transformers and 15 lines tripped in Sweden (Elovaara et al, 1992).
July 14: Railway signals were turned to red by the induced voltage (Wallerius, 1982)
A great magnetic storm occurred on 13 March 1989, that caused a nine-hour blackout of the 21,000 MW Hydro Québec, electric power system. A vivid description of that failure has been provided by G. Blais and P. Metsa(1993) of Hydro Québec:
Telluric currents induced by the storm created harmonic voltages and currents of considerable intensity on the La Grande network. Voltage asymmetry on the 735-kV network reached 15%. Within less than a minute, the seven La Grande network static var compensators on line tripped one after the other....With the loss of the last static var compensator, voltage dropped so drastically on the La Grande network (0.2 p.u.) that all five lines to Montréal tripped through loss of synchronism (virtual fault), and the entire network separated. The loss of 9,450 MW of generation provoked a very rapid drop in frequency at load-centre substations. Automatic underfrequency load-shedding controls functioned properly, but they are not designed for recovery from a generation loss equivalent to about half system load. The rest of the grid collapsed piece by piece in 25 seconds.
Many other power utilities in North America experienced problems ranging from minor voltage fluctuations to tripping out of lines and capacitors. A summary of these effects and the times of their occurrence is given by Cucchi and Ponder (1991).
Bozoki et al (1996) report other power system disturbances in North America in 1989:
On September 19, minor tank heating was observed on transformer T2 at Meadow Brook on the Allegheny power system and Public Service Electric & Gas: Salem #2, generator step-up transformer was damaged. On November 17-18, the Allegheny Power System Millvill SC tripped by neutral unbalance protection, and a BC Hydro 138 kV line tripped by overvoltage protection.
In Russia the geomagnetic storm affected automatic railway systems causing numerous signals to report false blockages of railway track of the Gorky railway system for several minutes at a time intermittently for as long as 3-4 hours (Eroshenko et al., 2010).
Protective relay operations occurred on power systems in North America during geomagnetic disturbances on March 22-24, April 28, May 16, June 4-5, June 10, and October 28-29. The details are given by Bozoki et al (1996).
During a period of strong geomagnetic activity on November 08-09 strong geomagnetically induced potentials were measured in Sweden's gas pipeline system (Lundstedt, 2006).
Solar eruptions on April 4, 2000 resulted in geomagnetic disturbances on April 6, 2000. Unexplained anomalies in railway circuits in Russia were reported (Eroshenko et al., 2010).
On July 15-16, railway circuits reported false blockages of railway tracks in Russia during a period of high geomagnetic activity over a four hour period (Eroshenko et al., 2010). The same storm caused the simultaneous tripping of capacitor banks in power grids in northern Alabama and northern Mississipi, USA (Kappenman, 2003).
On March 31, 2001 railway circuits reported false blockages of tracks in Russia for a period of high geomagnetic activity for a three hour period in the early morning and a one hour period in the afternoon (Eroshenko et al., 2010). A capacitor bank in a south eastern New York, USA power grid tripped early on March 31 (Kappenman, 2003).
A solar x-ray flare and halo coronal mass ejection that erupted on November 4, 2001 resulted in disturbed geomagnetic conditions on November 6, 2001. Geomagnetically induced currents reached peak values on many power grids (for example, in Japan), and problems were reported in New Zealand, including the destruction of a transformer (Béland and Small, 2004).
Damage to several transformers was reported in South Africa (Gaunt and Coetzee, 2007) in association with the Halloween storm, and a large geomagnetically induced current was reported in the Finnish natural gas pipeline (Pirjola et al., 2005). The city of Malmö in southern Sweden suffered from a power blackout that left 50,000 without power (Pulkkinen et al., 2005; Lundstedt, 2006).
In China, induced currents in the Halloween storm are responsible for vibration and noise in transformers at the Shanghe substation and the Ling’ao nuclear power plant (Liu et al., 2009). In Russia, unexplained anomalies in railway circuits were observed for several on both October 29th and 30th (Eroshenko et al., 2010).
A burst of solar activity in early November led to high levels of geomagnetic activity November 7-12, 2004. Geomagnetically induced currents were observed in power systems in Europe, Africa, North America, and China, and pipeline voltages exceeded normal operating ranges for long periods of time (Trichtchenko et al., 2007; Liu et al., 2009). Signals indicating false blockages of railway tracks in Russia were reported for numerous intervals November 8-10 (Eroshenko et al., 2010).>