Study of geomagnetic storms and assessment of their impact on technology and infrastructure in Spain and PortugalNatural risks
This article looks at geomagnetic storms and their effects on technological resources. It sets forth action guidelines that enable the most storm-vulnerable companies, institutions and public services to take emergency measures designed to avoid or reduce storm damage. The problem is studied in its broadest context: Space Weather. This helps everyone involved to build up a better understanding of the various storm-related physical phenomena and also helps to explain this threat to stakeholders and the public at large.
By MIGUEL HERRAIZ SARACHAGA. Doctor in Physics. Chair-holding professor of Earth Physics in the Departamento de Física de la Tierra, Astronomía y Astrofísica I, Universidad Complutense. GRACIA RODRÍGUEZ CADEROT. Doctor in Physics. Tenure-holding professor in the Departamento de Física de la Tierra, Astronomía y Astrofísica I, Sección Departamental de Astronomía y Geodesia in the Facultad de Matemáticas, Universidad Complutense. MARTA RODRÍGUEZ BOUZA. Physics graduate. MSc in geophysics and meteorology. PhD student. IZARRA RODRÍGUEZ BILBAO. Physics graduate. MSc in geophysics and meteorology. PhD student. FRANCISCO SÁNCHEZ DULCET. Physics graduate. MSc in geophysics and meteorology. Army officer and diploma holder in geodesy. Honorary collaborator of Universidad Complutense. BEATRIZ MORENO MONGE. Doctor in Mathematics. ICT researcher for the Centro de Innovación de Infraestructuras Inteligentes (Ci3). IRIA BLANCO CID. Physics graduate. MSc in geophysics and meteorology. PhD student. BENITO A. DE LA MORENA CARRETERO. Doctor in physics. Head of INTA’s Atmospheric Sounding Station in El Arenosillo, Huelva.
Geomagnetic storms, also called “magnetic storms” are worldwide disturbances in the Earth’s magnetic field, caused by solar-wind shock wave. Some geomagnetic storms have an intense effect on the Earth’s ionosphere, triggering off «ionospheric storms»; these are studied particularly closely in this article due to their effect on global navigation satellite systems (GNSSs). As will soon be explained in greater detail, geomagnetic storms stem from violent energy processes occurring in the Sun, going by the generic name of «solar storms». We are therefore dealing here with three types of storm: solar, ionospheric and geomagnetic. The three are closely bound up with each other but it is nonetheless useful to distinguish them clearly at the start.
Most geomagnetic storms are small and cause little damage. In the northern hemisphere they manifest themselves in the form of aurorae borealis, visible in high-latitude areas like Iceland, Greenland or North Norway, Sweden and Finland. In the southern hemisphere they produce a similar phenomenon called «aurora australis». Intense storms, however, like the ones analysed in this article, are also quite frequent events, causing not only the aurorae but also great damage. The US agency NOAA (National Oceanic and Atmospheric Administration) estimates that in any 11-year solar cycle there might be four “extreme”, 100 «severe» and 200 «strong» geomagnetic storms. These figures show we are dealing with a frequent natural phenomenon.
The first notion of the importance of very intense geomagnetic storms came on 1 and 2 September 1859, when a huge solar flare produced the biggest geomagnetic storm recorded to date. This event was dubbed the « Carrington Event» in honour of the English astronomer who observed the solar flare and related it to the magnetic storm recorded on Earth. The storm produced aurorae that were clearly visible on the Iberian Peninsula and even at latitudes close to 20º. It also damaged the telegraphy systems of the time, causing many pieces of equipment to burst into flame. This was the first known incident of this type affecting a technological resource and the event is now seen as a wake-up call about the possible influence that a storm of this type might have on our totally technology-dependent society.
The «Quebec Blackout» of March 1989 left 5 million people without electricity for 9 hours and caused 12 million dollars’ worth of transformer damage
Some idea of the important effects of a large geomagnetic storm can be gained by considering the resources most likely to be damaged, like satellites, power grids, gas- and oil-pipelines and air- and rail-transport. The effects on satellites could unleash a chain of knock-on effects on navigation, communication and positioning systems, triggering the collapse of systems as varied as air and sea traffic, security and surveillance and banking systems. In turn, the electric fields generated by the variations in the magnetic field during a geomagnetic storm are capable of inducing electric currents in conducting systems (power cables, metal conductions and earthed conductor systems, etc.). These currents, called Geomagnetically Induced Currents (GICs), pose a great danger to underground metal pipelines like oil- or gas-lines, and also to electricity systems. Damage to high-voltage power distribution systems, especially large transformers, could trip long-lasting and wide-ranging blackouts, damaging basic services like traffic lights, transport systems, water treatment systems and critical installations like hospitals, fire stations and nuclear power plants, with the concomitant nuisance and harm for millions of people in the affected areas. Witness the famous «Quebec Blackout» of March 1989, tripped by a magnetic storm. This power-cut left 5 million people without electricity for nine hours, caused transformer damage in the United States and Canada worth 12 million dollars and knocked out some of this same type of equipment in the United Kingdom. It also seriously affected many satellites. It is calculated that about 1600 orbiting satellites were temporarily out of control.
The likelihood of a storm like the « Carrington Event» reoccurring is currently a matter of hot debate. Riley has recently claimed there is a 12% chance of an event of this category occurring in the next 10 years, though many other scientists regard this probability as far-fetched. In 2013 Kataoka mooted a probability of 4-6%, still fairly high. Trustworthy estimates are very tricky due to the incomplete knowledge of some of the physical processes involved and absence of reliable data series to go on, but all studies do stress the reality of this natural peril.
If a Carrington-like magnetic storm should occur today, the consequences would be unimaginable, since technology dependence has soared since 1859. Estimates point to devastating cascade events in the US lasting several years and clocking up millions of dollars in losses. Odenwald and Green (2007), for example, have estimated an economic toll of 30 billion dollars due to the damage that any Carrington-like magnetic storm would do today to satellites in geostationary orbit. The impact could be so great that certain heavily industrialised countries like the United States and the United Kingdom have included this threat in a list of natural risks and have taken initiatives to head it off. Witness, among other examples, the recommendations published by such institutions as FEMA (Federal Emergency Management Agency) and NASA (National Aeronautics and Space Administration) in the USA and the National
Risk Register (NRR) of Civil Emergencies in the United Kingdom. This concern has been tabled in international organisations like the OECD (Organisation of Economic Cooperation and Development), which published in 2011the Geomagnetic Storms report.
Space Weather, a new discipline
Geomagnetic storms are a manifestation of Space Weather, a new field of study initiated back in the nineties of last century. It looks at the conditions of the Sun and solar wind, the magnetosphere, ionosphere and thermosphere, which might impact on the performance and dependability of technological systems both on the ground and in space and might even pose a threat to human health. Broadly speaking, the studies making up space weather take in three fields: the Sun and its atmosphere (as the source of the energy), interplanetary space (as the means of propagation) and the magnetosphere, ionosphere and surface of the Earth (as the effected regions). A better understanding of space weather and the design of early warning systems are crucial in the fight to mitigate risks associated with geomagnetic storms.
In Spain the social response to this problem is still very limited, although the Directorate General of Civil Protection (Dirección General de Protección Civil) organised conferences in 2011, 2012 and 2013 to explain the problem to institutions and companies most likely to be hit and affected by a strong storm. A particularly practical and timely initiative was carried out by the Regional Authority of Extremadura (Junta de Extremadura) in March 2011, when it published a Decalogue of Good Practices to Prevent Damage from a Severe Solar Storm (Decálogo de buenas prácticas. Tormenta Solar severa, cómo prevenir). Legislative initiatives include a Bill passed on this matter on 27 March 2012 by the Spanish lower house (Congreso de los Diputados), on a proposal made by the Socialist Group and driven by iniciativas ciudadanas (Citizen Initiatives). In the scientific community magnetic storms are now being studied from various angles in several universities (especially Complutense de Madrid, Complutense de Alcalá de Henares and Politécnica de Cataluña) and in other organisations such as Observatorio del Ebro (Ebro Observatory), the Instituto Geográfico Nacional (National Geographical Institute) and the Instituto Nacional de Técnica Aeroespacial (National Institute of Space Technology).
The overall context of the problem
Contradictory as it might seem, a magnetic storm on Earth begins in the Sun. The Sun acts on the Earth through its gravitational field, electromagnetic radiation (of which visible light and heat are the most everyday examples) and the continual emission of material from its corona, making up what is known as the «solar wind». This wind is a plasma flow of protons, electrons and alpha particles. This plasma is extraordinarily thin, with a density of only 10 particles per cm3 in the vicinity of the Earth. Under normal conditions this wind moves at a speed of about 400 km/s and drags along the Sun’s magnetic field with it. The characteristics of electromagnetic radiation and the solar wind vary greatly with the Sun’s activity level, which is both cyclical and sporadic in nature. Its cyclical nature is expressed in a period of about 11 years, called the solar cycle. Increasing activity is reflected by an increase in the number of sunspots and a higher number of violent outbreaks, which constitute the sporadic activity. Although unforeseeable, therefore, the sporadic behaviour also has a certain periodical character, tending to increase in number during phases when solar activity is nearing its peak.
Violent events in the Sun reach the Earth in the form of electromagnetic radiation with an 8-minute time lag and also in the form of particles and disturbances of the interplanetary magnetic field with a time lag of several hours to a few days
Sunspots are so called because they are visible as dark spots on the Sun’s surface, the photosphere. Their different colour from the rest of the Sun stems from their temperature, which, at about 4500 K, is lower than the c. 6000 K temperature elsewhere. The number of sunspots is measured by the «Wolf number», a widely used indicator to assess the Sun’s activity, with a reliable series of measurements dating right back to 1848. The magnetic field near the sunspots builds up to huge values and manifests itself violently in the form of gigantic eruptions known as solar flares. The areas with spots are therefore considered to be «active regions» of the Sun. Solar flares, together with Coronal Mass Ejection (CME), a closely related phenomenon, and coronal holes (regions with magnetic fields opening freely into the heliosphere), throw huge amounts of the corona’s mass into the interplanetary medium, modifying the speed and density of the solar wind.
Solar flares, for their part, are classified in terms of their peak X-ray flux, measured in W/m2. They are thus broken down into five main classes A, B, C, M and X. Each class is further broken down into a linear scale of 1 to 9, each number being twice as powerful as the former. Classes M and X indicate phenomena that might have important effects in near-Earth space. For example, the three most important solar flares of 13 and 14 May 2013 were generated by the same active region (cluster of sunspots AR 11748) and clocked up values of X1.7, X2.8 and X3.2.
On the initiative of Wolf (1816-1893) solar activity cycles are numbered successively from 1 onwards, 1 applying to the period 1755-1766. In this article we are paying particularly close attention to cycle 23, which lasted approximately from May 1996 to December 2007, and cycle 24 in which we are still immersed today.
Violent events in the Sun reach the Earth in the form of electromagnetic radiation with an 8-minute time lag and also in the form of particles and disturbances of the interplanetary magnetic field, dragged by the solar wind, with a time lag of several hours to a few days. This means that solar events have occurred eight minutes before their observation on Earth but there is then a time lag of several hours before the disturbances generated by these solar events affect the Earth and might trigger a geomagnetic storm. There is hence a chance of predicting the appearance of a geomagnetic storm and taking preventive action against it.
Solar wind emission is constant and its interaction with the Earth’s magnetic field produces a border where the forces cancel out, called the «magnetopause», which protects a cavity known as the «magnetosphere» (fig 1). This region of space where the Earth’s magnetic field exerts its influence is, so to speak, our home in the planetary system. The magnetosphere protects us from the action of the solar wind and cosmic rays, allowing life to exist in this safeguarded region.
Formation and measurement of geomagnetic storms
Geomagnetic storms are triggered by an increase in the plasma density and the speed of the solar wind after a solar flare or an earthwards-directed coronal mass ejection. These increases raise the pressure of the solar wind in the magnetopause and deform the magnetosphere. On the daytime side the magnetopause approaches our planet along the Sun-Earth line, moving in from 11 Earth radii to only 4-5. At the same time the region corresponding to the night-time hemisphere stretches out in a very complex manner, similar to a tube of toothpaste squeezed in the middle. This intensifies the Earth’s magnetic field and increases its bow-wave pressure against the solar wind, reaching a new equilibrium position. All these phenomena give rise to the geomagnetic storm, which affects, to a lesser or greater degree, the whole planet. Depending on the speed of the disturbed solar wind, it will occur between one and four days after the violent event on the surface of the Sun.
Not all coronal mass ejections produce magnetic storms on the Earth. In general, three conditions have to obtain for this to occur: (1) the solar storm has to be energetic enough, reaching either class X or high values of class M; (2) the coronal mass ejection has to be directed earthwards, meaning that the sunspot cluster or active region initiating the whole process must be on the visible side of the Sun away from its limbs; (3) the Bz component of the interplanetary magnetic field (IMF) dragged by the solar wind must be negative so that the lines of this field can join up with those of the Earth (reconnection of the IMF with the terrestrial field). It has recently been shown that IMF fluctuations, before their encounter with the magnetopause, are an important factor, little understood as yet, in whether the solar wind disturbance will trigger a geomagnetic storm.
The abovementioned conditions explain why an increase in solar activity might not necessarily be accompanied by an increase in geomagnetic storms. For example, the aforementioned solar flares of 13-14 May 2013 did not generate significant geomagnetic storms, since none of the ejections was sufficiently orientated towards the Earth.
A magnetic storm can be divided into three phases (fig. 2):
- Initial Phase. This is characterised by an increase in the density of field lines due to rising pressure of the solar wind. The pre-storm value of the horizontal component H of the Earth’s magnetic field then increases by a factor of between 30 and 50 nanoteslas (nT). This variation may last one or two hours although it does not occur in some storms.
- Main phase. During this phase the equatorial ring current is boosted by an injection of energised plasma. This occurs two to ten hours after storm commencement and may last several hours. There is a characteristic sharp fall in H.
- Recovery phase. This is the stage when the magnetic field returns to normal. It might last days.
Magnetic storms are measured against three geomagnetic scales; the most widely used ones are the Disturbance Storm Time (Dst) and three-hour indices. Dst is an index of magnetic activity obtained from four magnetometer stations near the equator and spread around the Earth’s perimeter. This index measures the magnetic field variation due to the equatorial ring current; it is calculated from the mean of the magnetic field’s horizontal component. The value of Dst is statistically zero on days considered to be calm by international organisations. During a magnetic storm the value falls for a few hours from zero to its minimum value and then slowly climbs back up to the initial value close to zero. Using this index, storms can be classified as shown in Table 1. Table 2 lists the geomagnetic storms that occurred in cycle 24 up to December 2013 with their corresponding Dst readings. A good idea of their size can be gained if we bear in mind that the «Carrington Event» recorded –850 nT while the Quebec storm weighed in at –640 nT.
|Category||Dst Value (nT)|
|Weak||-30 > Dst > -50|
|Moderate||-50 > Dst > -100|
|Intense||-100 > Dst|
The three-hour indices indicate geomagnetic activity in each of the last three-hour periods, thus providing eight readings a day. The main one is the K index, introduced by Bartels in 1938, giving a quantitative assessment of a magnetic disturbance linked to the Sun’s corpuscular emission. The data series was then extended back to 1932.
K is calculated using magnetograms, daily records of the magnetic field obtained from geomagnetic observatories.
From the magnetogram the H and D (declination) components are taken, cancelling out the magnetic variations due to the Sun in non-stormy conditions and the moon. The magnetogram is then divided into 8 three-hour intervals, the variation of H and D is measured and the biggest value provides the K index. The K scale varies from 0 to 9 and depends on the latitude, since the disturbance will vary in direct proportion to the nearness of the observatory to the auroral zones.
The Kp index is an indicator of planetary scope deriving from the K parameter. It is obtained as the mean value of standardised K indices in 13 observatories in the 44º-60º latitude belt in the northern or southern hemispheres. This index strikes a statistical relationship between the magnetosphere’s energy state and the five-level NOAA-rated size of the magnetic storms, represented by a G number (Table 3).
|Category||Kp Value||NOAA Scale|
Some international agencies take Kp 4 as the cut-off point for giving out warnings of a geomagnetic storm.
Geomagnetic storm study methodology: application of this methodology to the storm of 24-25 October 2011 and its impact on the Iberian Peninsula
With the aim of clearly presenting the various processes that trip a geomagnetic storm and its effects on the Earth in general and the Iberian Peninsula in general, a detailed monitoring is described below of the storm that broke out on 24-25 October 2011. This storm has been chosen for our example because it was intense (Dst = –132 nT), there is a lot of data to go on and it had a notable effect on the EGNOS (European Geoestationary Navigation Overlay Service), augmentation system, an essential resource for fine tuning GPS (Global Positioning Service) and Glonass (Globalnaya Navigatsionnaya Sputnikovaya Sistema) navigation systems in Europe and Africa. Some important effects of this storm on the Earth’s environment have been studied by Blanch et al.
The methodology studies successively the events on the Sun, the path of the solar wind to the Earth, its interaction with the Earth’s magnetic field, tripping the geomagnetic storm, the influence of this storm on the Earth’s ionosphere and the effect on EGNOS and air traffic safety.
Start of the event on the Sun
The whole process kicked off on 22 October 2011, with a solar eruption magnitude M1 that peaked at about 11:10 UT. This eruption produced a huge coronal mass ejection which disturbed the conditions of the solar wind. Figure 3, taken from the LASCO (Large Angle and Spectrometric Coronagraph) equipment onboard the SOHO satellite, shows the scale of the event. SOHO is located at Lagrange Point L1 where the gravitational attraction of the Sun and Earth cancel each other out; the orbits in this zone therefore have a higher gravitational stability. This zone lies 1,500,000 kilometres from the Earth.
Path of the CME towards the Earth
Figure 4 shows the forecast pathway through space of the CME generated by the solar flare. This has been drawn up from NOAA’s WSA-Enlilcone model. In the lefthand figure the CME, marked in red, is seen to be heading clearly for Mars, represented by a red dot top right; at this point it does not seem destined to impact on Earth. The central figure shows how, six hours later, it would reach Stereo A (small red square), an observatory orbiting the Sun and observing the star from two opposite positions of the same orbit, together with its twin observatory Stereo B, both of the satellites between them therefore giving a better idea of the structure and trend of solar storms. The forecast indicates that finally (righthand figure) the CME would end up brushing the Earth (represented by a yellow circle).
During the afternoon of 24 October (18:00 UT), the ACE satellite (Advanced Composition Explorer), also orbiting the Sun at Lagrange point L1, detected an increase in solar wind speed from 350 km/s to 550 km/s, indicating the incoming impact of the CME and auguring a geomagnetic storm (Fig.5). Furthermore the magnetic field’s Bz component was facing southwards, thus satisfying one of the aforementioned necessary conditions for a geometric storm being generated on Earth. The satellite was able to gauge the characteristics of the solar wind about 40 minutes before it hit the Earth. This provided a precious leeway for taking geomagnetic-storm mitigating measures.
In its journey earthwards, and at only 35,800 kilometres from its surface, the disturbance hit the geostationary satellites GOES 13 (longitude 75º) and 14 (longitude 135º), which were also able to assess the solar wind. Figure 6 shows the proton and electron flux readings for 24-27 October 2011, as recorded by GOES-13. The electron flux clearly shows the change caused by the disturbance as from 18.00 hours on the 24th and lasting until 9.00 hours on the 25th. The strong compression of the Earth’s magnetic field during the impact enabled the solar wind to penetrate deep into the magnetosphere from 19:06 UT until 19:11 UT, exposing the satellites to the action of the solar wind plasma.
Arrival of the solar wind on Earth. Recording of the storm on the Iberian Peninsula
At 18.00 hours (Universal Time) on 24 October 2011 the ACE satellite (Advanced Composition Explorer) detected an increase in solar wind speed, auguring a geomagnetic storm that hit the Earth about 40 minutes later
The arrival of the disturbed solar wind on the Earth generated a geomagnetic storm that was recorded in various observatories, affecting the ionosphere and throwing satellite positioning systems out of kilter. Figure 7 shows the magnetograms of 24 and 25 October obtained from the observatory of San Pablo de los Montes (Toledo). These show that the geomagnetic storm started on the afternoon of the 24th and lasted, at least, throughout the whole of the 25th. Figure 8, with the Dst index readings, shows that the initial phase of the geomagnetic storm lasted from 15 to 18.00 hours, the main phase lasted until 10.00 hours on the 25th and the recovery phase until 23.00 hours on the 29th.
Impact on the ionosphere and the Iberian Peninsula
The impact on the ionosphere (the conducting part of the atmosphere extending from 60 out to 2000 kilometres from the Earth’s surface) is crucially important for satellite communication purposes due to its strong influence on the transmission of electromagnetic waves. When the impact of the solar wind causes an appreciable variation in the ionosphere’s characteristics, then an «ionospheric storm» is said to have occurred. If this modification entails an electron-density increase (number of electrons per unit of volume) of the ionosphere, this is said to be a «positive ionospheric storm». Conversely, if the effect is a reduction in this density, it is said to be a «negative storm». In both cases there may be significant disturbances to GNSS communication systems.
Many specialist space weather stations issue a geomagnetic storm alert two days before it occurs
This impact has been studied in two different ways. Firstly by analysing ionograms (ionosphere readings from high frequency ionosondes) of 23, 24 and 25 October from the Observatorio del Ebro and INTA’s Estación de Sondeos Atmosféricos (Atmospheric Sounding Station) of the Centro de Experimentación (Experimentation Centre) in El Arenosillo (Cedea) Huelva. The results (Figure 9 shows those for El Arenosillo) reveal a slight increase in the critical frequency of layer F2, foF2, and a significant increase in the height of its maximum electron concentration, hmF2. The increase of foF2 tallies with the increase in density and will show up in the electron content analysis, since the plasma frequency is proportional to the square root of the electron density. The increase in height, for its part, is a characteristic phenomenon of geomagnetic storms whenever there is a positive ionospheric storm.
The second technique used for analysing the impact of the magnetic storm on the ionosphere was studying the variation in the total electron content (TEC). This parameter measures the number of electrons contained in a 1m2 cross-section cylinder running from the satellite to the receiver along the line of sight; its unit is called TECu and is tantamount to 1016 electrons/m2. The TEC reading is obtained from the delay in the transmission of electromagnetic waves observed in GPS stations. Its variations therefore show how the ionosphere has been affected by the geomagnetic storm, i.e., they gauge the importance of the generated ionospheric storm. In our study the analysis was made by processing the RINEX files (Receiver Independent Exchange Format) obtained in the stations shown in Figure 10. For each station a calculation was made of the vertical TEC, vTEC, and its relative value (expressed as vTECrel in this study), which is the difference of the value in each epoch divided by the monthly mean value of the days not disturbed magnetically.
The expression is:
where subscript i indicates the station under consideration.
Odenwald and Green (2007) have estimated an economic toll of 30 billion dollars due to the damage that any Carrington-like magnetic storm would do today to satellites in geostationary orbit
These results were then used as the basis for analysing the ionospheric storm phases and their relation to phases of the geomagnetic storm. A positive phase of the ionospheric storm is considered to exist when the difference between vTEC and the mean value is over 10 TECus or the relative vertical TEC tops 50%. Conversely, a negative phase of the ionospheric storm is considered to exist when the difference between vTEC and the mean value is less than 10 TECUs or the relative vertical TEC is less than –50%. Figure 11 gives the results obtained from a selection of the stations distributed by latitude. The readings thus obtained clearly show that the geomagnetic storm generated an ionospheric storm over the Iberian Peninsula. There is also a positive phase of the ionospheric storm corresponding to the initial phase of the geomagnetic storm, which is appreciable only at the lowest latitudes, and another of the same sign between the main phase and the start of the recovery phase of the geomagnetic storm. A series of four negative phases of the ionospheric storm also shows up during the recovery phase of the geomagnetic storm. In the two positive phases the maximum variation is heavily latitude-dependent. This effect also obtains in the duration of these phases, albeit with a less significant difference. In the negative phases the latitude effect is less notable.
Geomagnetic storms and the railway
Geomagnetic storms can also affect the railway system. The first ever mention of effects of this type came in the New York Times of 16 May 1921, when a news report linked the widespread failure of the signalling and control system of the New York Central Railroad and subsequent fire to a magnetic storm that produced visible aurorae in the region of New York. Geomagnetic storms were later blamed too for signalling failures in Sweden on 13-14 July 1982 and in Russia during many other magnetic storms. The explanation for these railway failures might be sudden voltage surges created by the storm-engendered geomagnetically induced currents (GICs). These voltage surges might upset the signalling system and confuse activation of free-line and occupied-line indications[3,4,5].
Close attention should be paid to this problem when considering the viability of high-speed railway lines (where safety measures have to be much stricter) for high-latitude countries like north Russia, Sweden, Norway and Finland.
- Odenwald, S. Newspaper reporting of space weather: End of a golden age. Space Weather, 2007, (5) S11005. doi:10.1029/2007SW000344.
- Wik, M; Pirjola, R; Lundstedt, H; Viljanen, A; Wintoft, P; Pulkkinen, A. Space weather events in July 1982 and October 2003 and the effects on geomagnetically induced currents on Swedish technical systems. Annales Geophysicae, 2009, (27) 1775-1787.
- Eroshenko, EA; Belov, AV; Boteler, D; Gaidash, SP; Lobkov, SL; Pirjola, R; Trichtchenko, L. Effects of strong geomagnetic storms on Northern railways in Russia. Adv. Space Res., 2010, (46) 1102-1110. doi:10.1016/j.asr.2010.05.017.
- Kasinskii, V; Ptitsyna, NG; Lyahov, NN; Tyasto, MI; Lucci, N. Effect of Geomagnetic Disturbances on the Operation of Railroad Automated Mechanisms and Telemechanics. Geomagnetism and Aeronomy, 2007, (47) 676-680. doi: 10.1134/S0016793207050179.
- Ptitsyna, NG; Kasinskii, VV; Villoresi, G; Lyahov, NN; Dorman, LI; Lucci, N. Geomagnetic effects on mid-latitude railways: A statistical study of anomalies in the operation of signaling and train control equipment on the East-Siberian Railway. Advances in Space Research, 2008, (42) 1510-1514. doi:10.1016/j.asr.2007.10.015.
To analyse in greater depth the disturbance caused on the Iberian Peninsula, a study was also made of the time trend of IPP-associated TEC readings. IPP stands for the ionospheric pierce point, i.e., the point where the satellite-receiver signal intersects with the ionosphere, which is assumed to be concentrated at a height of 350 kilometres. IPP maps are regional or global, representing the TEC readings at these points. To obtain them, a calculation is made first of the vTEC in the IPPs where the information is available. This calculation was made every minute on the days analysed for the stations of the Iberian Peninsula and with an average of five satellites for each epoch. From these readings a selection was made of those of most interest for this analysis, building up maps from them using Kriging interpolation with a 0.4º x 0.4º grid. The grid was chosen on the basis of several trials, the best results being obtained for the aforementioned distance. Figure 12 shows the maps corresponding to 11.00 hours of 24, 25 and 26 October 2011, at which time the vTEC variation reaches significant values. These maps clearly show the increase in vTEC produced in the ionospheric storm.
Effects on EGNOS services and positioning
The technological effect of greatest interest produced by this storm was the disturbance it caused to the EGNOS system, developed to improve the performance of Glonass, GPS and Galileo in Europe and Africa. The particular effect observed was a degradation of the APV-1 (Approach with Vertical Guidance) service, which ensures exact positioning with GNSS signals of 16 metres in the horizontal plane and 20 metres in the vertical plane. It includes two types of key information: HPL-VPL (Horizontal and Vertical Protection Levels) and HAL-VAL (Horizontal and Vertical Alert Limits).
Figure 13 shows the level of confidence on 23 October, with HPL and VPL falling within the alarm limits (HAL, VAL) for the satellite Egnos PRN120. A level of 99% is seen to reign almost throughout the whole of Europe and above 75% over the Iberian Peninsula. Due to the effect of the storm, this region shrank on the 24th and almost completely disappeared on the 25th (Fig. 14). On the 26th and 27th confidence levels from before the main phase of the storm were recovered, bearing out the effect of the geomagnetic storm.
Design of an alert protocol
In an effort to reduce the effects of geomagnetic storms, particular attention has been paid to early detection. The approach has been two-pronged. Firstly a research line has been set up to tap into the time sequence of the phenomena leading to the geomagnetic storm; this sequence has been painstakingly described for the storm of 24-25 October 2011. This is the approach followed by specialist space weather centres, many of which issue a storm warning with one or two days’ notice. This is therefore a predictive approach of great practical interest. The second approach centres on ionospheric effects and attempts to fend off disturbance once the ionospheric storm has begun. The objective here is not to predict the ionospheric storm but rather announce its occurrence soon enough to ascertain the failures that might be caused in communication, navigation and positioning systems.
The objective of the developed system is not to predict the ionospheric storm but rather to announce its occurrence soon enough to ascertain the failures that might be caused in communication, navigation and positioning systems
For this purpose the System for Rapid Information on Ionospheric Disturbances (Sistema de Información Rápida de Perturbaciones Ionosféricas) has been set up for the Iberian Peninsula and Southern Europe. This system is described below. The system is activated manually upon receiving an alert from one of the abovementioned centres. In the near future it will be tripped automatically. Once activated, the system stays in operation for 10 days to ensure the whole period of disturbances is studied. The program automatically carries out the following operations:
- Download the Rinex files and the navigation files from the 16 selected GNSS stations (Fig. 15) for the day of the possible alert and each one of the 10 previous days.
- Process the files using the abovementioned method to obtain the vTEC in each era, for each day and station.
- Calculate the mean vTEC of the 10 previous days and the vTECrel. These figures are stored in graph form for subsequent mining and review. vTECrel values are then filtered to remove wrong data that might trip the alert system erroneously. This filter consists of removing previous eras in which the vTECrel suffered brusque changes.
- Check the vTECrel value against the threshold value (±50%); the alert message is issued if the vTECrel exceeds the threshold value in at least 50% of the stations. This message then allows users to take due preventive measures.
The system is outlined in figure 16. To check out its validity, the information system has been applied to five storms that occurred in December 2006, October 2011, January 2012, April 2012 and July 2012. This crosschecking study took in 24 days around the date on which the Dst index recorded its minimum value, comparing this with the days on which an ionospheric storm alert message was issued. These tests gave a correct result in 77.18% of the days studied. Of the 29 days with an ionospheric storm, alert messages would have been issued on 23 of them; on only 21 of the 89 days without any disturbance would an unnecessary alert message have been issued. These results bear out system validity. In the near future this system will be speeded up and then made available to users on internet.
Geomagnetic storms are natural processes affecting the whole planet, inducing important physical phenomena like aurorae, radiation boosts, GICs and ionospheric storms. Intense geomagnetic storms might have a huge impact on many of the technological resources that now underpin our daily life. Satellites, power lines, navigation and railway systems may all by damaged by a big magnetic storm, generating huge economic losses and upsetting the workings of our society, creating a host of problems of unimaginable dimensions. There is therefore now a pressing need to pay this natural risk due attention and get across to the public at large the importance of readiness.
When put through its paces, the system gave a correct result on 77.18% of the days studied, vouching for its validity
Notable breakthroughs in space weather, some of which have been presented in this article, enable us nowadays to give about 30-40 minutes’ notice of the arrival of a solar wind disturbance on Earth, possibly sparking off a geomagnetic storm. This advanced notice gives society the chance to take measures to protect both people and material goods, providing it has been properly trained up to do so beforehand. Awareness of this threat needs to be raised by government authorities, institutions, educators and the media; this article aims to contribute to this awareness-raising effort, which now needs to be backed up by an ongoing informative effort.
- González, WD; Joselyn, JA; Kamide, Y; Kroehl, HW; Rostoker, G; Tsurutani, BT; Vasyliunas, VM. What is a geomagnetic storm? Journal of Geophysical Research, 1994, (99) 5771-5792.
- Carrington, RC. Description of a singular appearance seen in the Sun on September 1. Mon. Not. Roy. Astron. Soc. 1859, (20) 13-15./li>
- Hodgson, R. On a curious appearance seen in the Sun. Mon. Not. Roy. Astron. Soc. 1859, (20) 15-16.
- Green, J; Boardsen, S. Duration and extentof the great auroral storm of 1859. Adv. Space Res. 2006. (38) 130-135.
- Boteler, DH. The super storms of August/September 1859 and their effects on the telegraph system. Adv. Space Res., 2006, (38) 159-172.
- Hapgood, MA. Towards a Scientific Understanding of the Risk from Extreme Space Weather. Adv. Space Res., 2006, doi:10.1016/j.asr.2010.02.007.
- Riley, P. On the probability of occurrence of extreme space weather events, Space Weather, 2012, (10), doi: 10.1029/2011sw000734.
- Kataoka, R. Probability of occurrence of extreme magnetic storms, Space Weather, 2013, (11) doi:101002/swe.20044.
- Odenwald, S; Green, J; Taylor, W. Forecasting the impact of an 1859-calibre superstorm on satellite resources. Adv. Space. Res. 2007, (38) 280-297.
- Aguado, J. Recuperación de la magnetosfera terrestre tras sucesos solares geoefectivos. Tesis doctoral, Universidad de Alcalá, 2010.
- Buonsanto, MJ. Ionospheric storms-a review, Space Science Reviews, 1999, 88: 563–601.
- Parkinson, W.D. Introduction to Geomagnetism, Scottish Academic Press, 1983.
- Siscoe, G; NU. Crooker and CR. Clauer, Dst of the Carrington storm of 1859, Adv. Space. Res. 2006, (38) 173-179./li>
- Lakhina, GS; S. Alex; BT. Tsurutani y WD. González. Research on Historical Records of Geomagnetic Storms, Proceedings IAU Symposium No. 226, 2005 International Astronomical Union, doi:10.1017/S1743921305000074.
- Blanch, E; Marsal, S; Segarra, A; Torta, JA; Altadill, D; Curto, JJ. Space weather effects on Earth ́s environment associated to the 24-25 October 2011 geomagnetic storm. Space Weather, 2013, (11) 153-168.
- Kelley, M. The Earth’s Ionosphere: Plasma Physics and Electrodynamics, 2nd ed., 2009, Academic Press, Amsterdam.
- Mendillo, M; Storms in the ionosphere: Patterns and processes for total electron content, Rev. Geophys, 2006, 44, 2 RG4001, doi:10.1029/2005RG000193.
- Saiz, E; C. Cid; Y. Cerrato, Forecasting intense geomagnetic activity using interplanetary magnetic field data. Annales Geophysicae. 2008, (26), 39893998.
- Cerrato, Y., E. Saiz, C. Cid, 2012, Terrestrial Magnetosphere, in Advances in Solar and Solar-Terrestrial Physics, Editors: Georgeta Maris and Crisan Demetrescu,177-206, ISBN: 978-81-3080483-5.