Severe Space Weather Events: Understanding Societal and Economic Impacts: A Workshop Report (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Introduction Historical Background As evidenced in both ancient legend and the historical record, human activities, institutions, and technologies have always been prey to the extremes of weather—to droughts and floods, ice storms and blizzards, hurricanes and tornados. Around the middle of the 19th century, however, society in the developed parts of the world became vulnerable to a different kind of extreme weather as well—to severe disturbances of the upper atmosphere and the near-Earth space environment driven by the magnetic activity of the Sun. Although the nature of the solar- terrestrial connection was not understood at the time, such disturbances were quickly recognized as the culprit behind the widespread disruptions that periodically plagued the newly established and rapidly expanding telegraph networks. During the following century and a half, with the growth of the electric power industry, the develop- ment of telephone and radio communications, and a growing dependence on space-based communications and navigation systems, the vulnerability of modern society and its technological infrastructure to “space weather” has increased dramatically. The adverse effects of extreme space weather on modern technology—power grid outages, high-frequency communication blackouts, interference with Global Positioning System (GPS) navigation signals, spacecraft anomalies—are well known and well documented. The physical processes underlying space weather are also gener- ally well understood, although our ability to forecast extreme events remains in its infancy. Less well documented and understood, however, are the potential economic and societal impacts of the disruption of critical technologi- cal systems by severe space weather. Defining and quantifying these impacts presents a number of questions and challenges with respect to the gathering of the necessary data, the methodology for assessing the risks of severe space weather disturbances as low-frequency/high-consequence events, the perception of risk on the part of policy makers and stakeholders, and the development of appropriate risk management strategies. As a first step toward charting the dimensions of the problem of determining the socioeconomic impacts of extreme space weather events and addressing the questions of space weather risk assessment and management, a public workshop was held on May 22-23, 2008, in Washington, D.C., under the auspices of the National Research Council’s (NRC’s) Space Studies Board. The workshop brought together representatives of industry, the govern- ment, and academia (attendees are listed in Appendix B) to consider both direct and collateral effects of severe space weather events, the current state of the space weather services infrastructure in the United States, the needs of users of space weather data and services, and the ramifications of future technological developments for contem-  

INTRODUCTION  porary society’s vulnerability to space weather. The workshop concluded with a discussion of “the way forward,” in which the participants identified un- or underexplored topics relevant to the question of space weather impacts, highlighted various weaknesses in the existing space weather services infrastructure, and suggested improvements that would yield the greatest benefits in space weather risk management. The key themes, ideas, and insights that emerged during the workshop’s 1½ days of informative presentations and lively discussions are summarized in this report, which was prepared by the members of the ad hoc NRC Com- mittee on the Societal and Economic Impacts of Severe Space Weather Events: A Workshop tasked with organizing the workshop (Appendix D). To set the stage for the chapters that follow, we begin with a description of the mag- netic superstorms of August-September 1859, by some measures the most severe space weather event on record. Known as the Carrington event, the 1859 storms were referred to throughout the workshop as an example of the kind of extreme space weather event that, if it were to occur today, could have profound societal and economic consequences, with cascading effects throughout the complex and interrelated infrastructures of modern society. The Great Magnetic Storms of August-September 1859 (the Carrington Event) Shortly after midnight on September 2, 1859, campers in the Rocky Mountains were awakened by an “auroral light, so bright that one could easily read common print.” The campers’ account, published in The Rocky Mountain News, continues, “Some of the party insisted that it was daylight and began the preparation of breakfast.” 1 Eighteen hundred miles to the east, Henry C. Perkins, a respected physician in Newburyport, Massachusetts, observed “a perfect dome of alternate red and green streamers” over New England. To the citizens of Havana, Cuba, the sky that night “appeared stained with blood and in a state of general conflagration” (Figure 1.1). Dramatic auroral displays had been seen five nights before as well, on the night of August 28/29, when (again in the words of Dr. Perkins) “the whole celestial vault was glowing with streamers, crimson, yellow, and white, gathered into waving brilliant folds.”2 In New York City, thousands gathered on sidewalks and rooftops to watch “the heavens . . . arrayed in a drapery more gorgeous than they have been for years.” The aurora that New Yorkers witnessed that Sunday night, The New York Times assured its readers, “will be referred to hereafter among the events which occur but once or twice in a lifetime.”3 From August 28 through September 4, auroral displays of extraordinary brilliance were observed throughout North and South America, Europe, Asia, and Australia, and were seen as far south as Hawaii, the Caribbean, and Central America in the Northern Hemisphere and in the Southern Hemisphere as far north as Santiago, Chile (Figure 1.2).4 Even after daybreak, when the aurora was no longer visible, its presence continued to be felt through the effect of the auroral currents. Magnetic observatories recorded disturbances in Earth’s field so extreme that magnetometer traces were driven off scale, and telegraph networks around the world—the “Victorian Internet” 5—experienced major disruptions and outages. “The electricity which attended this beautiful phenomenon took possession of the magnetic wires throughout the country,” the Philadelphia Evening Bulletin reported, “and there were numerous side displays in the telegraph offices where fantastical and unreadable messages came through the instruments, and where the atmospheric fireworks assumed shape and substance in brilliant sparks.” 6 In several locations, operators disconnected their systems from the batteries and sent messages using only the current induced by the aurora. 7 The auroras were the visible manifestation of two intense magnetic storms that occurred near the peak of the sunspot cycle. On September 1, the day before the onset of the second storm, Richard Carrington, a British amateur astronomer, observed an outburst of “two patches of intensely bright and white light” 8 from a large and complex group of sunspots near the center of the Sun’s disk. The outburst lasted 5 minutes and was also observed, indepen- dently, by Richard Hodgson from his home observatory near London. Carrington noted that the solar outburst—a white-light flare—was followed the next day by a magnetic storm, but he cautioned against inferring a causal connection between the two events. “One swallow,” he is reported to have said, “does not make a summer.” 9 Space Weather: “The Mysterious Connection Between the Solar Spots and Terrestrial Magnetism” The dazzling auroral displays, magnetic disturbances, and disruptions of the telegraph network that occurred between August 28 and September 4, 1859, were recognized by contemporary observers—at least the scientifically 

 SEVERE SPACE WEATHER EVENTS—UNDERSTANDING SOCIETAL AND ECONOMIC IMPACTS FIGURE 1.1  “The red light was so vivid that the roofs of the houses and the leaves of the trees appeared as if covered with blood” (report of the aurora seen in San Salvador, September 2, 1859; see note 2 at the end of this chapter). Low-latitude red auroras, such as those widely reported to have been observed during the Carrington event, are a characteristic feature of major geomagnetic storms. The aurora shown here was photographed over Napa Valley, California, during the magnetic storm of November 5, 2001. Reprinted with permission from D. Obudzinski (www.borealis2000.com). © Dirk Obudzinski 2001. FIGURE 1.2  Locations of reported auroral observationsGreen.eps ~1.5 hours of the September 2, 1859, magnetic storm 1.2 during the first (orange dots). Courtesy J.L. Green, NASA bitmap 

INTRODUCTION  informed among them—as especially spectacular manifestations of a “mysterious connection between the solar spots and terrestrial magnetism.”10 This connection had been established earlier in the decade on the basis of the regular correspondence observed between changes in Earth’s magnetic field and the number of sunspots. 11 Well- established by this time as well was the “intimate and constant connection between the phenomena of the aurora borealis and terrestrial magnetism.”12 And by the mid-1860s, Hermann Fritz in Zürich and Elias Loomis at Yale would furnish convincing evidence of a link between the occurrence of the aurora and the sunspot cycle. 13 “We must therefore conclude,” Loomis wrote in Harper’s New Monthly Magazine, “that these three phenomena—the solar spots, the mean daily range of the magnetic needle, and the frequency of auroras—are somehow dependent the one upon the other, or all are dependent upon a common cause.” 14 Although the existence of the link among solar, geomagnetic, and auroral phenomena was recognized by the time of the 1859 events, the nature of this link was not understood. The white-light flare observations by Carrington and Hodgson furnished a critical clue. But it would not be until the 1930s that the significance of their observations was appreciated, and a full picture of the phenomena that constitute what we now call “space weather” would not emerge until well into the space age.15 A major turning point in our understanding of space weather came with the discovery of coronal mass ejections (CMEs) in the 1970s and with the recognition that these, rather than eruptive flares, are the cause of non-recurrent geomagnetic storms.16 Large-scale eruptions of plasma and magnetic fields from the Sun’s corona, CMEs contain as much as 1016 grams or more of coronal material and travel at speeds as high as 3000 kilometers/second, with a kinetic energy of up to 1032 ergs.17 Eruptive flares and CMEs occur most often around solar maximum and result from the release of energy stored in the Sun’s magnetic field. CMEs and flares can occur independently of one another; however, both are generally observed at the start of a space weather event that leads to a large magnetic storm. To be maximally geoeffective, i.e., to drive a magnetic storm, a CME must (1) be launched from near the center of the Sun onto a trajectory that will cause it to impact Earth’s magnetic field; (2) be fast (≥1000 kilometers/second) and massive, thus possessing large kinetic energy; and (3) have a strong magnetic field whose orientation is opposite that of Earth’s.18 The cause of the magnetic storm that began on September 2, 1859, was thus not the highly energetic flare19 that Carrington and Hodgson had observed the previous morning. It was a fast CME launched from or near the same giant sunspot region just northwest of the Sun’s center that had produced the flare. Had the Solar and Heliospheric Observatory (SOHO) been in operation in 1859, its Large-Angle and Spectrometric Coronagraph (LASCO) would have observed the CME some 20 minutes or so after the flare’s peak emission at 11:15 GMT. The CME would have appeared as a bright “halo” of material surrounding the occulted solar disk, indicating that it was headed directly toward Earth (Figure 1.3). Between the time of the flare/CME eruption on September 1 and the onset of the magnetic storm the next morning, 17 hours and 35 minutes elapsed. 20 Dividing the mean distance between Earth and the Sun by the 17.5-hour propagation time yields a speed of approximately 2300 kilometers per second, making the CME of September 1, 1859, the second fastest CME on record. 21 Moving substantially faster than the surrounding medium, fast CMEs create a shock wave that accelerates coronal and solar wind ions (predominantly protons) and electrons to relativistic and near-relativistic velocities. Particles are accelerated by solar flares as well; and large solar energetic particle (SEP) events, although dominated by shock-accelerated particles, generally include flare-accelerated particles (some of which may be further accel- erated by the shock). Traveling near the speed of light, SEPs begin arriving at Earth within less than hour of the CME lift-off/flare eruption and are channeled along geomagnetic field lines into the upper atmosphere above the North and South poles, where they enhance the ionization of the lower ionosphere over the entire polar regions— polar cap absorption (PCA) events—and can initiate ozone-depleting chemistry in the middle atmosphere. 22 SEP events—“solar radiation storms” in NOAA terminology—can last several days. 23 The mid-19th century lacked the means to detect and measure SEPs, and its most sophisticated technologies were unaffected by them. Thus, in contrast to the widely observed auroral displays and magnetic disturbances, the radiation storm unleashed by the solar eruption on September 1 went unnoticed and undocumented by contem- porary observers. There is, however, a natural record of the storm that can be retrieved and interpreted. Nitrates, produced by SEP bombardment of the atmosphere above the poles, settle out of the atmosphere within weeks of a SEP event and are preserved in the polar ice. Analysis of anomalous nitrate concentrations in ice core samples 

10 SEVERE SPACE WEATHER EVENTS—UNDERSTANDING SOCIETAL AND ECONOMIC IMPACTS 1.3 hires.eps FIGURE 1.3  An X17 flare observed during the 2003 “Halloween” storms with SOHO’s Extreme-ultraviolet Imaging Tele- scope (EIT) (left) and a difference image showing the associated halo CME (right). SOHO is stationed 1.5 million kilometers bitmap upstream from Earth, at the Lagrangian point 1. These images suggest what might have been observed on September 1, 1859, if 19th-century technology had been capable of building a SOHO-like space-based solar observatory. Courtesy NASA/ESA. allows the magnitude of historical—i.e., pre-space-era—SEP events to be estimated. 24 Such an analysis indicates that the 1859 event is the largest SEP event known, with a total fluence of 1.9 × 1010 cm–2 for protons with ener- gies greater than 30 MeV, four times that of the August 1972 event. 25 The shock responsible for the radiation storm hit Earth’s magnetosphere 26 at 0450 GMT on September 2. It dramatically compressed the geomagnetic field, producing a steep increase in the magnitude of the field’s hori- zontal (H) component,27 which marked the onset of the geomagnetic storm. The compression of the field would also have triggered an almost instantaneous brightening of the entire auroral oval (Figure 1.4). The CME arrived shortly after the passage of the shock and triggered the main phase of the storm, the severity of which can be inferred from contemporary reports of low-latitude auroras and magnetometer data from the Colaba Observatory in Bombay, India.28 The equatorward boundary of the aurora moves to increasingly lower latitudes (relative to its nominal location at 55°-65° magnetic latitude) with increasing storm intensity. 29 The observations of the aurora as far south as the West Indies, Jamaica, Cuba, and San Salvador are thus evidence that the September storm was extraordinarily intense. A rough quantitative measure of its intensity is provided by the Colaba data, which show a precipitous reduction (1600 nT) in H at the peak of the storm’s main phase. Converted to 1-hour averages, these data yield a proxy Dst index of approximately –850 nT.30 For comparison, the largest Dst index recorded since the International Geophysical Year (1957) is –548 nT for the superstorm of March 14, 1989. 31 Without upstream solar wind measurements such as are provided today by the Advanced Composition Explorer, researchers can only speculate about the structure of the CME and the magnitude and precise orienta- tion of the associated magnetic fields.32 What can be inferred with certainty from the intensity and duration of the September storm, however, is that very strong magnetic fields were associated with the CME and that their orientation was opposite that of Earth’s. This allowed the two fields to merge and enormous amounts of energy to be transferred into the magnetosphere, producing the magnetospheric and ionospheric phenomena characteristic of a major magnetic storm: (1) increased earthward flow of magnetospheric plasma, creating or intensifying the ring current;33 (2) the explosive release of stored magnetic energy in multiple magnetospheric substorms; (3) an increase in the energy content of the radiation belts as well as the possible creation of temporary new belts; (4) the 

INTRODUCTION 11 FIGURE 1.4  Far-ultraviolet images of the pre-shock1.4 hires.eps (right) aurora obtained with the auroral imager on (left) and post-shock bitmap NASA’s IMAGE satellite during the July 14-15, 2000, “Bastille Day” event. Courtesy NASA/IMAGE FUV team. development of intense auroral currents (electrojets) in the upper atmosphere; and (5) changes in the ionospheric and thermospheric density at midlatitudes. The storm was at its most intense on September 2, and the geomagnetic field required several days to recover. Balfour Stewart, the director of the Kew Observatory near London, reported that the magnetic elements “remained in a state of considerable disturbance until September 5, and scarcely attained their normal state even on September 7 or 8.”34 The same chain of events described for the September storm—CME/eruptive flare onset, SEP acceleration (probable), impact of the shock/CME on Earth’s magnetic field, the resulting magnetospheric and ionospheric disturbances—will also have occurred in the case of the August 28/29 storm. The occurrence of low-latitude auroras and the dramatic auroral displays witnessed at higher latitudes indicate that this was a severe storm as well, although recently analyzed data from Russian magnetic observatories show that it was less intense and of shorter duration than the September 2 storm.35 No solar eruptions were reported in association with the August event, and so the transit time or shock/CME speed cannot be determined. It is not known whether the CME was SEP-effective as well as geoeffective.36 However, it is not unreasonable to speculate that a less intense SEP event was associated with the August 28/29 storm.37 Space Weather Effects and Socioeconomic Impacts The August-September auroral and magnetic storms of 1859 were recognized by contemporaries as extraordi- nary events, and they still rank at or near the top of the lists of particularly severe geomagnetic storms. 38 Given the state of technology in the mid-19th century, their societal impact was limited to the disruptions of telegraph service “at the busy season when the telegraph is more than usually required,”39 the telegraph companies’ associated loss of income, and whatever the attendant effects on commerce and railroad traffic control might have been. 40 Today the story is quite different. Modern society depends heavily on a variety of technologies that are vul- nerable to the effects of intense geomagnetic storms and solar energetic particle events. Strong auroral currents, which wreaked havoc with the telegraph networks during the Carrington event, can disrupt and damage electric power grids and may contribute to the corrosion of oil and gas pipelines. Magnetic storm-driven ionospheric density disturbances interfere with high-frequency (HF), very-high-frequency (VHF), and ultra-high-frequency (UHF) radio communications and navigation signals from GPS satellites. Exposure of spacecraft to energetic particles during SEP events and radiation belt enhancements can cause temporary operational anomalies, damage 

12 SEVERE SPACE WEATHER EVENTS—UNDERSTANDING SOCIETAL AND ECONOMIC IMPACTS FIGURE 1.5  LASCO images from October 28, 2003, showing the effect of solar energetic particle bombardment on one of the SOHO coronagraphs. The image on the left is LASCO CME. The image on the right was obtained ~8.5 hours later. 1.5 of the halo large.eps Courtesy NASA/ESA. bitmap critical electronics,41 degrade solar arrays, and blind optical systems such as imagers and star trackers (Figure 1.5). Moreover, intense SEP events present a significant radiation hazard for astronauts on the International Space Station during the high-latitude segment of its orbit as well as for future human explorers of the Moon and Mars who will be unprotected by Earth’s magnetic field.42 In addition to such direct effects as spacecraft anomalies or power grid outages, a complete picture of the impact of severe space weather events on contemporary society, with its complex weave of dependencies and interdependencies, must include the collateral effects of space-weather-driven technology failures. For example, polar cap absorption events can degrade—and, during severe events, completely black out—HF communications along transpolar aviation routes, requiring aircraft flying these routes to be diverted to lower latitudes, at a not inconsiderable cost to the airlines43 and inconvenience to the passengers. WORKSHOP PLANNING AND REPORT STRUCTURE This workshop report was prepared by the members of the committee responsible for organizing the May 2008 workshop. In response to its statement of task (Appendix A), the Committee on the Societal and Economic Impacts of Severe Space Weather Events: A Workshop held a planning meeting prior to the workshop at which it gathered information on the issues to be explored. During and following that meeting the committee developed and refined the workshop structure, identified appropriate speaker candidates, and developed targeted questions and other materials for the speakers and sessions. The workshop consisted of eight topical sessions, each with a moderator, a rapporteur, and a panel of speakers representing different stakeholder industries, organizations, and agencies (see the workshop agenda in Appendix B). There were two summary sessions as well, plus a brief intro- ductory talk by Daniel Baker, director of the Laboratory for Atmospheric and Space Physics at the University of Colorado and chair of the committee. Each panelist received a separate set of questions intended to elicit infor- mation relevant to the goals outlined in the committee’s statement of task. That information is summarized in the succeeding chapters. The structure of the report follows, with one exception, the order of the topical sessions, with each chapter summarizing the key points made during the panelists’ presentations and the subsequent discussions and summary sessions. The exception is the session on extreme space weather events, held on the second day of 

INTRODUCTION 13 the workshop. That session’s presentation by Jim Green (NASA Headquarters) on the Carrington event serves as the starting point for the discussion of 1859 storms in this introductory chapter, while Paul O’Brien’s (Aerospace Corporation) presentation on planning for extremes and extreme value analysis is summarized in Chapter 7, “Future Solutions, Vulnerabilities, and Risks.” Abstracts were received from most of the workshop speakers, and those are included, as submitted, in Appendix C of this report. The majority of the figures included in this report were taken from the presentations made by the workshop panelists. NOTES   1. Quoted in Green, J.L., et al., Eyewitness reports of the great auroral storm of 1859, Adv. Space Res. 38, 145-153, 2006, p. 149. This is one of a collection of papers published in a special issue of Advances in Space Research dedicated to the August-September 1859 geomagnetic storms. Extensive use was made of this collection in the preparation of this introduction, as reflected in the notes that follow. Popular accounts of the Carrington event can be found in Clark, S., The Sun Kings: The Unexpected Tragedy of Richard Carrington and the Tale of How Modern Astronomy Began, Princeton University Press, Princ- eton, N.J., 2007, and Odenwald, S., and J.L. Green, Bracing the satellite infrastructure for a solar storm, Scientific American, August 2008.   2. Shea, M.A., and D.F. Smart, Compendium of the eight articles on the “Carrington Event” attributed to or written by Elias Loomis in the American Journal of Science, 1859-1861, Adv. Space Res. 38, 313-385, 2006, p. 149. Elias Loomis (1811- 1889) was a professor of natural philosophy at Yale University with a particular interest in meteorology. Loomis collected reports of the aurora and magnetic disturbances observed during the 1859 storms and published them in eight installments in the American Journal of Science. These were compiled by Shea and Smart and published in the special issue of Advances in Space Research referred to in note 1. Henry Perkins’ report is contained in Loomis’ third article and appears on pp. 332-333 of the ASR compendium; the description of the red aurora seen over Havana is from a report published in the first installment; it appears on p. 326 of the ASR compendium.   3. The New York Times, August 30, 1859.   4. Green, J.L., and S. Boardsen, Duration and extent of the great auroral storm of 1859, Adv. Space Res. 38, 130-135, 2006; Cliver, E.W., and L. Svalgaard, The 1859 solar-terrestrial disturbance and the current limits of extreme space weather activity, Solar Physics 224, 407-422, 2004. Cliver and Svalgaard (p. 419, Table VII) rank the aurora of September 2 second on the list of the six documented lowest-latitude auroras, after the great aurora of February 1872 (low-latitude extent = 19°); according to Green and Boardsen, however, the September 2 aurora extended to 18° geomagnetic latitude.   5. Standage, Thomas, The Victorian Internet: The Remarkable Story of the Telegraph and the Nineteenth Century’s On-Line Pioneers, Walker & Co., 1998.   6. The Philadelphia Evening Bulletin is quoted in The New York Times of August 30, 1859. Sparking started fires in some telegraph offices, and one operator, Frederick Royce of Washington, D.C., received “a very severe electric shock, which stunned me for a moment.” A witness saw “a spark of fire jump from [Royce’s] forehead to the sounder.” Royce’s account of his experience was reported in The New York Times of September 5, 1859, and reprinted by Loomis (note 2) and G.B. Prescott (note 7).   7. Prescott, G.B., History, Theory, and Practice of the Electric Telegraph, Ticknor and Fields, Boston, 1860, p. 320.   8. Carrington, R.C., Description of a singular appearance seen in the Sun on September 1, 1859, Mon. Not. Roy. Astron. Soc. 20, 13-14, 1860. Quoted in Bartels, J., Solar eruptions and their ionospheric effects—a classical observation and its new interpretation, Terr. Mag. 42, 235-239, 1937.   9. Carrington quoted in E.W. Cliver, The 1859 space weather event: Then and now, Adv. Space Res. 38, 119-129, 2006. The quote appears on p. 123. 10. Kirkwood, D., Solar phenomena, New Englander and Yale Review 19, 51-63, 1861, p. 62. 11. See Cliver, 2006, pp. 120-121, on the independent discovery in the early 1850s of the connection between geomag- netic activity and the number of sunspots by Edward Sabine, R. Wolf, and A. Gautier. 12. Prescott, G.B., The aurora borealis, The Atlantic Monthly: A Magazine of Literature, Art, and Politics 4, 740-751, 1859, p. 748. This article is incorporated almost verbatim in Prescott’s 1860 book on the telegraph (note 7). 13. Schröder, W., Herman Fritz and the foundation of auroral research, Planet. Space Sci. 46, 461-463, 1998. 14. Loomis, E., The aurora borealis or polar light, Harper’s New Monthly Magazine 39, 1-21, 1869. 15. See Cliver, 2006, pp. 124-127, on the interpretation of the Carrington event in the 1930s and the development of the modern understanding of solar-terrestrial relations. 16. Gosling, J.T., The solar flare myth, J. Geophys. Res. 98, 18937-18949, 1993. 17. Gopalswamy, N., Coronal mass ejections of solar cycle 23, J. Astrophys. Astron. 27, 243-254, 2006. 

14 SEVERE SPACE WEATHER EVENTS—UNDERSTANDING SOCIETAL AND ECONOMIC IMPACTS 18. Gopalswamy, 2006. 19. According to a conservative estimate of its intensity, “the Carrington flare was a >X10 soft x-ray event, placing it among the top ~100 flares of the last ~150 years.” See Cliver and Svalgaard, 2004, p. 410. 20. Bartels, 1937. 21. See Gopalswamy, 2006, p. 251, Figure 6. 22. Jackman, C.H., et al., Satellite measurements of middle atmospheric impacts by solar proton events in solar cycle 23, Space Sci. Rev. 125, 381-391, 2006. 23. For example, the largest >10 MeV SEP event of solar cycle 23 lasted 51/2 days, from 1705 UT on November 4, 2001, until 0715 UT on November 10, 2001. (See Report of Solar and Geophysical Activity for November 10, 2001, issued jointly by NOAA and the USAF.) 24. McCracken, K.G., et al., Solar cosmic ray events for the period 1561-1994. 1. Identification in polar ice, 1561-1950, J. Geophys. Res. 106, 21585-21598, 2001. 25. McCracken, 2001; Shea, M.A., et al., Solar proton events for 450 years: The Carrington event in perspective, Adv. Space Res. 38, 232-238, 2006. Shea et al. give a >30 MeV proton fluence of 5.0 × 109 cm–2 for the August 1972 SEP event (Table 1). They state that this was the “first major large solar proton fluence event that was recorded by a spacecraft” and “it is this event against which most comparisons are made” (p. 236). It should be noted that their Table 1 also includes the SEP event of November 12, 1960, for which a fluence twice that of the August event is given (9 × 10 9 cm–2). However, as Shea and Smart note in an earlier paper, there is considerable uncertainty about the actual value of the >30 MeV proton fluence during this event (Shea, M.A., and D.F. Smart, A summary of major solar proton events, Solar Physics 127, 297-320, 1990). For example, Kim et al. note that values as small as 1.3 × 109 cm–2 have been estimated for the November 1960 event (Kim, M.-H., X. Hu, and F.A. Cucinotta, Effect of shielding materials from SPEs on the lunar and Mars surface, paper presented at the AIAA Space 2005 Conference, August 30–September 1, 2005, AIAA 2005-6653, 2005). 26. The magnetosphere is the region of space dominated by the geomagnetic field. It is populated by electrically charged particles of varying composition (but mostly protons) originating in the solar wind and the ionosphere. The interaction with the solar wind stretches the magnetosphere on the anti-sunward side into a long, comet-like tail that can extend millions of miles downstream in the solar wind flow. 27. Cf. the magnetometer data from the Kew Observatory outside London, reproduced in Cliver, 2006, p. 123, Figure 4. 28. Tsurutani, B.T., et al., The extreme magnetic storm of 1-2 September 1859, J. Geophys. Res. 108(A7), 2003, doi:10.1029/2002JA009504. 29. Yokoyama, N., Y. Kamide, and H. Miyaoka, The size of the aurora belt during magnetic storms, Ann. Geophys. 16, 566-583, 1998. 30. Siscoe, G., N.U. Crooker, and C.R. Clauer, Dst of the Carrington storm of 1859, Adv. Space Res. 38, 173-179, 2006. The hourly Dst (disturbed storm time) index is the standard measure of magnetic storm intensity. It is derived from measure- ments made at four low-latitude magnetic observatories of the depression in the magnitude of the horizontal component of the geomagnetic field. The depression in the field is caused by an increase in the energy density of the ring current, a current system encircling Earth at low latitudes. It is the formation of a ring current that constitutes a magnetic storm. Use of the Colaba data for a Dst proxy assumes that the contribution of low-latitude auroral electrojects to the depression in H was insignificant. (For the opposite view, see Green and Boardsen, 2006, p. 134). It should be noted that Dst estimates for the September storm calculated on the basis of assumed solar wind parameters can yield higher values. Tsurutani et al., 2003, predict a Dst of –1760 nT. See also Li, X., et al., Modeling of the September 1-2, 1859, super magnetic storm, Adv. Space Res. 38, 273-279, 2006. In contrast, the upper limit Dst that Siscoe et al. derive from solar wind conditions is consistent with the proxy Dst of –850 nT. 31. Cliver and Svalgaard, 2004, p. 416, Table VI; Tsurutani et al., 2003. 32. According to Tsurutani et al., 2003, the storm had a single, brief (1-1.5 hrs) main phase and was caused by a magnetic cloud-type CME with an intense southward magnetic field and no contribution from a draped field in the sheath of shocked solar wind between the CME and the shock. Siscoe et al., 2006, on the other hand, hypothesize that the storm consisted of two main phases separated by a brief recovery. The first main phase was caused by a strongly southward sheath field; the second, by a northward-to-southward rotation of the field within the CME. 33. See note 30. 34. Stewart, B., On the great magnetic disturbance which extended from August 28 to September 7, 1859, as recorded by photography at the Kew Observatory, Phil. Trans. Royal Soc. 151, 423-430, 1861. 35. Nevanlinna, H., On geomagnetic variations during the August-September storms of 1859, Adv. Space Res. 42, 171- 180, 2008. 

INTRODUCTION 15 36. On the properties of SEP-effective shocks, see Gopalswamy, 2006, p. 250, §4.2 and Figure 5. 37. Smart, D.F., M.A. Shea, and K.G. McCracken, The Carrington event: Possible solar proton intensity-time profile, Adv. Space Res. 38, 215-225, 2006. 38. Cliver and Svalgaard (note 4) rank the Carrington event against other severe storms in terms of sudden ionospheric disturbance, SEP fluence, CME transit time, storm intensity, and equatorward extent of the aurora. They conclude, “While the 1859 event has close rivals or superiors in each of the above categories of space weather activity, it is the only documented event of the last ~150 years at or near the top of all the lists,” p. 407. 39. Walker, C.V., On magnetic storms and currents, Phil. Trans. Royal Soc. 151, 89-131, 1861. The quote is from p. 95: “The fact appears to have been that the disturbance was of such magnitude and of so long continuance, and this at the busy season when the telegraph is more than usually required, that our clerks were at their wits’ end to clear off the telegrams (which accumulated in their hands) by other less affected but less direct routes.” 40. Green et al., 2006, pp. 151-152, estimate a total global loss to the telegraph companies of $300,000 (lost revenue + operator labor loss) but note that there are not enough data to allow an estimate of the collateral impact of the telegraph outages. 41. Damage to Nozomi’s communications and power subsystems during a SEP event on April 21, 2002, contributed to the eventual loss of the Japanese Mars mission. The MARIE instrument on NASA’s Mars Odyssey is believed to have been irreparably damaged by SEP bombardment during the 2003 Halloween storms (Lee, K.T., et al., MARIE solar quiet time flux measurements of H and He ions below 300 MeV/n, 29th International Cosmic Ray Conference, 101-104, 2005). Ironically, MARIE was designed to measure the martian space radiation environment. 42. NRC, Space Radiation Hazards and the Vision for Space Radiation: Report of a Workshop, The National Academies Press, Washington D.C., 2006; NRC, Managing Space Radiation Risk in the New Era of Space Exploration, The National Academies Press, Washington, D.C., 2008. 43. “A typical flight duration for a polar route from a North American destination to Asia is over 15 hours. If the flight must divert for any reason, an additional stop-off is required. This results in considerable time loss, additional fuel, and the added time will require a whole new crew. The average cost of this kind of diversion is approximately $100,000.” NOAA, Intense Space Weather Storms October 19-November 07, 2003, NOAA National Weather Service, Silver Spring, Md., April 2004, p. 17.