Extreme Solar Flares As Drivers of Space Weather

Home , Heliophysics

Proposal to ISSI for an International Team in Space Science ”Team on probabilities of extreme solar ﬂares and CMEs” Karel Schrijver and J¨urg Beer 2011

Extreme solar ﬂares as drivers of space weather: from science towards reliable statistics

C.J. Schrijver Lockheed Martin Adv. Techn. Ctr., 3251 Hanover St., Palo Alto, CA 94304, U.S.

J. Beer EAWAG, P.O. Box 611, Uberlandstrasse¨ 33, D¨ubendorf 8600, Switzerland.

Abstract. The most energetic space weather events are most dangerous to our society that increasingly depends on space-based assets and the stable availability of electrical power and communication/navigation systems 1. Learning about such events is, however, difficult because they are also relatively rare. Advancing technologies enable us to obtain the probabilities and properties of extreme geospace weather other than by direct observation. Ice cores appear to contain information on extreme geospace weather in Earth’s past, while observations of stars like the Sun provide a statistical sample of many thousands of stars instead of the single one that we live with. The continuing analysis of multiple ice cores, and the ongoing monitoring of 150,000 stars by the Kepler satellite, for example, provide a wealth of information waiting to be utilized. We propose to assemble an international team of experts in the analysis of records in ice and rock, solar and stellar observations, and of the processes that link solar events to their observables to (1) set state-of-the-art bounds to the probabilities of the occurrence of space weather events of different magnitudes, and (2) to assess the potential of improving these constraints by identifying gaps in our knowledge or necessary future measurements and techniques to improve our knowledge. The international team brings together experts in solar and stellar flares, the analysis of paleo-chemical and paleo-radionuclide records from ice cores and rocks (from Earth, moon, and of meteoric origin), and solar energetic particles and their generation, propagation, and impact on elements of geospace and on other bodies in the heliosphere (including lunar rocks brought back during the Apollo era as well as meteorites). The primary output of the team, following two 4-day meetings in the second half of 2011 and the first quarter of 2012, is a joint review paper in which what we know about the most energetic solar events is outlined, together with what still needs to be uncovered. The team will aim to also summarize and publish findings that identify pathways to enhanced knowledge about extreme space weather in each of the sub-disciplines: solar physics, as- trophysics, geophysics, and atmospheric physics, as well as analysis of rock and ice-core samples. The final publication is anticipated to be ready for submission no later than the summer of 2012. The international team will, if this proposal is accepted, seek to add two young researchers working in heliophysics and in geo-sciences.

1See the NRC report on “Severe Space Weather Events – Understanding Societal and Economic Impacts” at http://www.nap.edu/catalog.php?record id=12507; and a high-level editorial on impacts of space weather: http://www.nytimes.com/2011/03/11/opinion/11iht-edholdren11.html? r=2&src=recg 0 Proposal to ISSI: Team on properties of extreme solar ﬂares and CMEs 1

Figure 1. Downward-cumulative flare frequency distribution for energies exceeding E, normalized to approximate solar-maximum X-ray flux density levels (1.3 × 105 erg cm−2 s−1; Judge et al. (2003)) assuming a linear dependence of the X-ray surface flux density. Flare spectra for Sun-like G-type stars are shown in black, for warmer and cooler stars in grey. The grey dashed power-law fit has an index of α f + 1 = −0.87. EUV data from Aschwanden (2000), soft X-ray data from Shimizu (1997), hard X-ray data > 8keV from Lin et al. (2001), and stellar data from Audard et al. (2000). The grey histograms for solar data bracket a conservative energy uncertainty of a factor of 2. Three estimates of flare energies for GOES X flares are shown near the top, from Aschwanden & Alexander (2001) and Benz (2008).

1. Rationale and goals Solar flares and associated coronal mass ejections drive space-weather events in geospace. Such events have already had significant societal and economic impacts, including major electrical power outages, disruptions in communications, problems in global positioning, and spacecraft anomalies and failures: although rare, the damage that may be inflicted to our global society and its safety and economy by extreme events is of such a magnitude that in-depth study of their properties and likelihood is prudent. Space weather events with substantial impacts on large segments of our society are, fortunately, infrequent. Since the beginning of regular X-ray monitoring of the Sun in 1976, for example, only 22 flares of class X10 or larger have occurred (only three of which were 2 C.J. Schrijver and J. Beer of class X20 or larger); extreme geomagnetic events (scale G5) occur only a few times per solar cycle. The downside of having such low-frequency high-impact events is, however, that we do not know the statistics of the most extreme space weather events well enough to successfully quantify the potential impact of such events by following the properties of the driving events through our technological infrastructure and from there into our economy. Our observational records of the Sun only provide a very limited view of what its magnetic activity has on offer primarily because our Sun exhibits extreme events at intervals that are long compared to researchers’ careers as well as to the era of advanced technology that has aided us in our observations. Consequently, we can expect even the ’present-day Sun’ to have surprises in store for us that we have not yet observed simply because we have not been looking long enough. Some of these surprises may lie hidden in records such as polar ice sheets, while others lie embedded in rocks from outer space. Observations of stars of similar mass and luminosity as the Sun provide interesting, but potentially disturbing information on the most extreme solar coronal storms that may occur: Sun-like stars have been observed to produce flares that are 1,000 times more energetic than an X1 flare, while some ’cool stars’ even produce flares up to five orders of magnitude larger than an X1 flare. Such large flares are readily observable in, e.g., the broad-band visible observations made with the Kepler satellite (see Walkowicz et al. (2011) for an initial assessment on a sample of 23,000 cool - i.e., with convective envelopes - main-sequence stars). Fig. 1 summarizes the results of solar and stellar flare studies from the pre-Kepler literature. Solar flares range from about 1032 ergs for X-class flares down to the detection threshold, currently some eight orders of magnitude lower. Flares with lower total energies emit typically most of their photons at longer wavelengths; the figure shows that low-energy flares are seen with EUV instrumentation, larger ones in soft X rays (SXT), and the largest flares in hard X rays (HXR). It may be that the X-ray and EUV flares do not form a continuous power-law distribution because of the changing sensitivity of the instruments as the spectral distribution of the photons changes with flare magnitude. On the other hand, the X-ray flares are invariably from active regions, while the EUV statistics are for mostly quiet- Sun conditions. As assessment of whether a true continuous power-law spectrum holds for all solar flares remains to be completed; there are already indications that the power law approximation breaks down for solar flares larger than about X10, with the present-day Sun exhibiting such large flares less frequently than expected from the power law approximation to data from more active stars in Fig. fig:flaredist (see the discussion by Schrijver (2011)2). Establishing where this breakdown occurs will be one of the team’s tasks. Stellar flare observations suggest power-law spectra with a similar slope to that found for the solar flares, but with the frequency of flares increasing with (in fact, more or less in proportion to) the quiescent coronal brightness. When the observations of the active stars for which flares are most readily observed are normalized by the quiescent coronal emission, the frequency distributions appear to form a continuation of the solar flare-energy spectrum, particularly for those stars with spectral types closest to that of the Sun (black histograms on the lower right of Fig 1). In assessing this compilation of solar and stellar flare observations (for the most recent workshop of Cool Stars, Stellar Systems, and the Sun), Schrijver (2011)2 points out an in- consistency between the observational record and the suggestion from stellar observations that the flare energy spectrum continues its power-law behavior well above X1 flares: “The

2Preprint at http://www.lmsal.com/∼schryver/Public/ms/cs16schrijver.pdf Proposal to ISSI: Team on properties of extreme solar flares and CMEs 3 frequency distribution in Fig. 1 suggests that if we estimate an X1 flare to have an energy of about 1032 ergs, we should see about 30 flares per year of that magnitude or larger during active phases in the solar cycle. That number compares quite well with the average frequency of 25 per year for X-class flares over the past three solar cycles, when counting only during the active half of the cycles (see the compilation by Hudson 2007). We would expect an X10 or larger about 4 times per year, which is high by a factor of about three given the 21 observed X-class flares since 1976 with the Sun in an active state for about 15 years within that interval. Extending that spectrum even further, we would expect an X100 flare or larger once every other year for the Sun near cycle maximum. As we have not experienced such large flares in at least half a century, we face the possibility that despite the intriguing align- ment of solar and stellar flare data after normalization to the mean coronal brightness, there may in fact be an upper limit to stellar flare energies that may shift to lower values [. . . ]: the solar flare energy distribution appears to drop below the power-law fit in Fig. 1 above X10, and the largest flare observed to date has been estimated to be an X45±5 (Thomson et al. 2004).” The primary goal of the proposed ISSI science team will be to improve our knowledge of the most energetic solar flares possible for the present-day Sun by combining information from solar, stellar, ice-core and rock-surface-layer sources. We propose that this be achieved by addressing the following tasks during two team meetings: Meeting I.

1. Review statistics of direct observations of solar and stellar magnetic storms, including photon and energetic-particle spectra. 2. Establish cross-calibration of diagnostics in various wavelength bands (and their uncertainties) to yield distributions as a function of total energy for both solar and stellar data. 3. Assess what kind of information we can expect from paleo-data (cosmogenic radionu- clides, nitrate, anything else), and define criteria which must be fulfilled to accept a specific kind of data as a reliable proxy for large space-weather events. 4. Review statistics of records in ice cores and rock surface layers of major solar energetic particle events. 5. Identify areas of missing information or major uncertainties in the steps to be taken to assemble the various observations into a coherent result for estimated bolometric flare energies for solar flares similar to that show in Fig. 1. 6. Identify and assign action items for the 2nd meeting, including any new ideas that emerged during the meeting not reflected in this proposal. Meeting II.

1. Assess consistency of solar, stellar, ice-core, and rock-face data. 2. Derive information on transport and impact processes from solar flares, through he- liospheric ICMEs, to storage and analysis in, primarily, snow/ice deposits in polar regions. 3. Complete a consensus study for the refereed literature on the outcome of the study team on the energy spectrum for solar flares. 4. Develop findings based on the study team’s activities that identify and, where possible, prioritize research areas in which – and methodologies by which – advances appear feasible and necessary to better constrain our knowledge of major solar events and their impacts on geospace. 4 C.J. Schrijver and J. Beer

To achieve these goals, a number of areas will need to be addressed. These problem areas guided us in assembling the expertise within the team: (1) address how to combine the various solar and stellar observations onto a single scale of ’characteristic flare energies’ that combine observations in the visible, UV, EUV, and X-rays; (2) assemble from the literature the available information on solar and stellar flares; (3) assess what information from Earth’s paleo-records is pertinent to solar flares and associated energetic particle events (including cosmogenic isotopes in ice and rocks as well as chemicals such as the so-called “odd nitrogen” constituents of ice cores) and how significant such constraints are; (4) assemble the statistical information into a compound best-estimate of flare energies for the present-day Sun and, if feasible, use stellar statistics to constrain statistics for the Sun throughout its evolution as activity subsided due to rotational braking. These problems echo issues identified in reviews that touch upon the proposed team effort by, e.g., Güdel (2007), Usoskin (2008) and Schrijver (2011). 2. Project output The anticipated output of the team includes at least a joint review paper in a high-quality peer-reviewed journal (possibly lthe Astrophysical Journal or the Journal of Geophysical Research), in which the findings of the team are represented as a consensus that identifies what is known and what still needs to be learned, along with the main uncertainties on the energy distribution of solar flares with particular emphasis on the largest, least frequent events. 3. Added value of ISSI ISSI provides a uniquely supportive environment for international teams to meet. The meeting facilities for groups of about a dozen members are ideally conducive for discussions, during the formal sessions as well as during breaks. Having a team meet in the compact city of Bern also means that there are ample opportunities for more detailed and focused discussions during breaks, dinners, or at other times that the team members are focused on the problem at hand and away from the many distractions they would have at their home institutions. The support for local expenses facilitates the attendance of the researchers by reducing the total cost of the trips significantly. The support staff at ISSI is highly competent and dedicated to the success of ISSI in general and the assembled teams in particular; internet support, help with presentational equipment, the means of exchange of presentations and other information all make the ISSI environment highly conducive to a successful outcome of the team effort.

Table 1. Team members (all have committed to participation) and their areas of expertise. Name Expertise Country Carolus Schrijver (lead) Solar-stellar activity USA Jürg Beer (co-lead) Radio-nuclides, ice-core analyses CH Urs Baltensperger atmospheric aerosols CH Ed Cliver Origin and propagation of solar energetic particles USA Hugh Hudson Solar flare observations and spectral properties USA Manuel Güdel Activity of Sun-like cool stars; flare scaling laws A Ken McCracken Fingerprinting of solar energetic events in geo-records AUS Thomas Peter atmospheric chemistry CH Rachel Osten Multi-spectral observations of stellar flares USA David Soderblom Cool-star flares in Kepler observations USA Ilya Usoskin Historical solar variability; atmo. effects of energetic particles SF Eric Wolff Nitrate in ice cores and the influences of solar particle events UK Proposal to ISSI: Team on properties of extreme solar flares and CMEs 5

4. Team members The international team members committed to supporting this proposal and their areas of expertise are listed in Table 1. 5. Schedule We propose for the team to meet twice, in meetings of 4-days each, with an interval of approximately 6 months between them. Ideally, the first meeting would take place in the late summer or early fall of 2011 (late August or early September, ISSI-schedules permitting) and the second around February or March of 2012. The team would aim to submit its final report and primary publication shortly after the second meeting, say in May or June of 2012. We plan to request the team to come to the first meeting with written summaries of facts, unknowns, and uncertainties on their area of expertise, armed with ideas of what to do to integrate the knowledge into a complete picture of the energy distribution of large to extreme flares on the Sun and on Sun-like stars. During the first meeting, the experts with a wide variety of backgrounds will exchange information, discussing in particular the assumptions going into analyses that others on the team might shed light on, leading to a mutual comprehensive understanding of the potential and limitations of the observations and models used to interpret them. The second half of the meeting will be used (a) to outline in substantial detail the joint primary publication and (b) to identify, compile, and prioritize action items in preparation for the second meeting. During the second meeting, the focus will be on presentations and discussions on all of the action items identified in the first meeting, and on reviewing and revising written texts and figures to assemble a first draft of the paper, and to advance in defining the conclusions as far as possible to take advantage of the available face-to-face discussions. In the 2-3 months after the second meeting, the team leaders will produce a full version of the joint paper, iterate with the team towards a consensus paper, and submit this to the peer-reviewed journal identified as the most suitable by the team. The team leaders will stimulate the publication of any supporting detailed papers and will formulate a document with recommendations and priorities for future research on extreme space weather events; a suitable publication venue for this will be identified during one of the two team meetings in consultation with the team. If this proposal is accepted, we shall seek two young researchers to join us, one likely with interests in combining solar and stellar data onto a common energy scale, and one with interests in the analysis and interpretation of paleo-records (either chemical are cosmogenic isotopes). The team leaders will not be requesting any travel reimbursement, and proposes to use this in favor of (full or partial) support of the young researchers. 6. Required facilities The team will require only a meeting room capable of holding about a dozen people, pro- jection equipment, and wireless internet access. 7. Requested financial support Only the standard financial support of the team members’ local expenses is requested. Re- imbursement of the leader’s travel expenses will be waived in favor of support for the local expenses (per diem) of two young researchers to be identified upon acceptance of this proposal. Publication charges will be covered by the institute of the lead proposer. 6 C.J. Schrijver and J. Beer

References Aschwanden, M. J. 2000, Solar Phys., 190, 233 Aschwanden, M. J., & Alexander, D. 2001, Solar Phys., 204, 91 Audard, M., Güdel, M., Drake, J. J., & Kashyap, V. L. 2000, ApJ, 541, 396 Benz, A. O. 2008, Living Reviews in Solar Physics, 5, 1 Güdel, M. 2007, Living Reviews in Solar Physics, 4, 3. 0712.1763 Hudson, H. S. 2007, ApJL, 663, 45. arXiv:0707.1118 Judge, P. G., Solomon, S. C., & Ayres, T. R. 2003, ApJ, 593, 534 Lemen, J. R., Title, A. M., Akin, D. J., Boerner, P., Chou, C., Drake, J. F., Duncan, D. W., Edwards, C. G., Friedlaender,F. M., Heyman, G. F., Hurlburt, N. E., Katz, N. L., Kusner, G., Levay, M., Lindgren, R. W., Mathur, D. P., McFeaters, E. L., Mitchell, S., Rehse, R. A., Schrijver, C. J., Springer, L. A., Tarbell, T. D., Wuelser, J.-P., Wolfson, C. J., Yanari, C., Bookbinder, J. A., Cheimets, P. N., Caldwell, D., Gates, R., Golub, L., Park, S., Podgorski, W. A., Scherrer, P. H., Bush, R. I., Gummin, M. A., Soufli, R., Windt, D. L., Beardsley, S., Clapp, M., Lang, J., & Waltham, N. 2010, Solar Phys. Lin, R. P., Feffer, P. T., & Schwartz, R. A. 2001, ApJL, 557, L125 Schrijver, C., & Siscoe, G. 2010a, Heliophysics. Evolving solar activity and the climates of space and Earth (Cambridge, U.K.: Cambridge University Press) — 2010b, Heliophysics. Space storms and radiation: Causes and Effects (Cambridge, U.K.: Cam- bridge University Press) Schrijver, C. J. 2001, ApJ, 547, 475 Schrijver, C. J. 2009, Advances in Space Research, 43, 739. 0811.0787 — 2010, ApJ, 710, 1480. 0912.0969 — 2011, in Cool Stars, Stellar Systems, and the Sun (16), edited by C. Johns-Krull, A. West, & M. Browning (San Francisco: Astron. Soc. of the Pacific). In press Schrijver, C. J., & Title, A. M. 2010, JGR Shimizu, T. 1997, in Magnetic Reconnection in the Solar Atmosphere, edited by R. D. Bentley & J. T. Mariska, vol. 111 of Astronomical Society of the Pacific Conference Series, 59 Thomson, N. R., Rodger, C. J., & Dowden, R. L. 2004, Geophys. Res. Lett., 31, 6803 Usoskin, I. G. 2008, Living Reviews in Solar Physics, 5, 3. 0810.3972 Walkowicz, L. M., Basri, G., Batalha, N., Gilliland, R. L., Jenkins, J., Borucki, W. J., Koch, D., Caldwell, D., Dupree, A. K., Latham, D. W., Meibom, S., Howell, S., Brown, T. M., & Bryson, S. 2011, AJ, 141, 50. 1008.0853 Wolk, S. J., Harnden, F. R., Jr., Flaccomio, E., Micela, G., Favata, F., Shang, H., & Feigelson, E. D. 2005, ApJSS, 160, 423. arXiv:astro-ph/0507151 Woods, T. N., Eparvier, F., Jones, A. R., Hock, R., Chamberlin, P. C., Klimchuk, J. A., Didkovsky, L., Judge, D., Mariska, J., Tobiska, W. K., Schrijver, C. J., & Webb, D. F. 2010 Proposal to ISSI: Team on properties of extreme solar flares and CMEs 7

Appendix 1: Contact information

J¨urg Beer Eawag, Swiss Federal Institute of Aquatic Science and Technology Postfach 611 CH-8600 Duebendorf (t) +41-58 765 51 11 (f) +41-58 765 5210 (e) [email protected]

Urs Baltensperger Paul Scherrer Institut Labor f¨ur Atmosph’arenchemie´ CH-5232 Villigen PSI SWITZERLAND (t) +41-56 310 2408 (e) [email protected]

Ed Cliver Space Weather Center of Excellence, Space Vehicles Directorate Air Force Research Laboratory Hanscom AFB, MA 01731-3010 USA (t) +1-781 377 3975 (f) +1-781 377 3160 (e) [email protected]

Hugh Hudson Space Sciences Laboratory University of California Berkeley, CA 94720 USA (t) +1-510-643-0333 (e) [email protected]

Manuel G¨udel University of Vienna Department of Astronomy T¨urkenschanzstr. 17 A-1180 Vienna Austria (t) +43-1 4277 53814 (f) +43-1 4277 9518 (e) [email protected]

Ken McCracken 100 Mt. Jellore Lane Woodlands, NSW, 2575 Australia (t) +61 2 4878 5121 (e) [email protected]

Thomas Peter ETH Zürich Institut für Atmosphäre und Klima CHN O 12.1 Universitätstrasse 16 8092 Zürich SWITZERLAND (t) +41-44 633 27 56 8 C.J. Schrijver and J. Beer

(e) [email protected]

Karel Schrijver Lockheed Martin Advanced Technology Center 3251 Hanover Street Palo Alto, CA 94304 USA (t) +1-650 424-2907 (f) +1-650 354 5445 (e) [email protected]

Rachel Osten Space Telescope Science Institute 3700 San Martin Drive Baltimore, MD 21218 USA (t) +1-410 338 4762 (f) +1-410 338 5090 (e) [email protected]

Dave Soderblom Space Telescope Science Institute 3700 San Martin Drive Baltimore, MD 21218 USA (t) +1-410 338 4543 (e) [email protected]

Ilya Usoskin Sodankyl¨aGeophysical Observatory (Oulu unit) P.O. Box 3000, FIN-90014 University of Oulu FINLAND (t) +358-8 553 1377 (f) +358-8 553 1390 (e) ilya.usoskin@oulu.ﬁ

Eric Wolﬀ British Antarctic Survey High Cross Madingley Road Cambridge CB3 0ET UK (t) +44 1223 221491 (e) [email protected] Proposal to ISSI: Team on properties of extreme solar ﬂares and CMEs 9

Appendix 2: Curricula vitae