1. ISSI Jan 2009 Early Stages of Eruptions Alphonse C. Sterling 1 NASA/MSFC 1 Currently at JAXA/ISAS, Sagamihara, Japan

2. ISSI Jan 2009 Introduction What happens at the start of solar eruptions? –Preflare phase. –Hints about the trigger mechanism? We can’t see the coronal field, but have been attempting to use filaments as tracers of the erupting field. “Slow rise phase” --> Flare+“fast rise phase” We have examined several individual events, but more still needed!

3. ISSI Jan 2009 Quiet-Region vs. AR Eruptions Some aspects of quiet-region events: Larger scale (filaments of up to a few 10 5 km). More slowly evolving (pre-eruptive motions typically several hours). OK for analysis with EIT (FOV = full disk; cadence = 12 min). Easy to find in EIT. (STEREO will be better, of course. Possible cavity connection. If we assume quiet-region eruptions and active-region eruptions are essentially the same, except for (basically) magnetic-field strength, can learn about all eruptions from quiet-region events (cf. Sterling & Moore 2005; Williams et al. 2005). AR eruptions: Faster, more energetic. Easier to identify heatings resulting from reconnections, etc. Study AR eruptions when possible (TRACE, Hinode, etc.)

4. ISSI Jan 2009 First, an Example

5. ISSI Jan 2009Sterling, Moore, Thompson (2001); Sterling & Moore (2004a)

6. ISSI Jan 2009 Sterling, Moore, Thompson (2001)

7. ISSI Jan 2009 Some Theories for Eruption Onset: Tether cutting. Breakout model. MHD Instability. (Generally, this applies to “fast eruption” onset, although slow rise might be part of the process.) These theories are testable (at least to within some limits) by our observations.

8. ISSI Jan 2009 Some Theories for Eruption Onset: Tether cutting. Moore & Labonte (1980); Sturrock (1989); Moore et al. (1997; 2001). (Also Chen & Shibata 2000; Lin et al. 2001; van Ballegooijen & Martens 1989 (?))  Fundamentally bipolar.  Energy release via reconnection deep inside the “core field.” Breakout model. Antiochos (1998); Antiochos et al. (1999).  Fundamentally multi-polar, with bipole core fields and restraining overlying fields.  Earliest energy release via slow reconnection at interface. MHD Instability. Sturrock et al (2001); Rust & LaBonte (2005); Fan & Gibson (2007); etc.  Rapid rise prior to reconnection onset.

9. ISSI Jan 2009 (Moore et al. 2001)

10. ISSI Jan 2009 Runaway Tether-Cutting Reconnection (Moore & Sterling 2006)

11. ISSI Jan 2009 Breakout Reconnection Moore & Sterling 2006

12. ISSI Jan 2009 Ideal MHD Instability Moore & Sterling 2006

13. ISSI Jan 2009 First example, revisited:

14. ISSI Jan 2009Sterling, Moore, Thompson (2001); Sterling & Moore (2004a)

15. ISSI Jan 2009

18. Cavity Examples

19. ISSI Jan 2009

25. HXT Lo 14 – 23 keV

26. ISSI Jan 2009 Perhaps tether cutting has trouble explaining observations of this event. In general however, fast rise and onset of brightenings in SXRs occur in close synchrony. Breakout signatures can occur earlier, but this is not definitive. Do we see the chewy n?

27. ISSI Jan 2009

31. Observed Characteristics Sample size. So far, we have examined about 12 (including 2 AR) events “in detail” (e.g., motions and intensity changes); mapped trajectories of about 25 additional events. Trajectory. Filament eruptions often undergo two stages: slow and fast (e.g., Tandberg-Hanssen et al. 1980, Bong et al. 2006; cf. Ohyama & Shibata 1997). Slow-rise linearity. In several events, slow rise is fit better with line than with a polynomial or exponential. (Slow-rise is sometimes complex, however; Akiyama & Sterling.) Flaring at start of fast eruption. Onset of SXR and HXR flaring coincides ``closely’’ with start of fast eruption. Breakout signatures. Occur close to time of start of fast eruption. (Moore & Sterling 2006; Bong et al. 2006)

32. ISSI Jan 2009 Bong et al. (2006) Breakout after start of fast eruption(?)

33. ISSI Jan 2009 Observed Characteristics – Cont. Dimmings. (Discussed by many workers; hard to generalize, e.g., Howard & Harrison 2004.) –Local dimmings. Intensity dimmings next to and along neutral line begin weakly during slow-rise phase (stretching of field lines), and become strong dimmings at start of fast-rise phase. –Remote dimmings (and brightenings). In some cases, dimmings and brightenings occur at locations far removed from main neutral line, consistent with breakout-type reconnection. Magnetic cavities. These show that filaments belong to a more extended magnetic environment; interaction with overlying fields can lead to remote brightenings/dimmings. (Cf. Gibson et al. 2006.)

34. ISSI Jan 2009 We have made progress on the fast-eruption- onset question, but issue is still not settled. Hard to say for sure signatures for which mechanism “come first.” (And perhaps some operate in combination.) An equally fundamental question however, is: What triggers the slow rise? This question may be easier to address.

35. ISSI Jan 2009 An AR-event example from Hinode On-disk filament eruption of 2 March 2007, seen with TRACE, STEREO. Hinode: –SOT (FG V magnetogram), etc. –SXRs from XRT Also use MDI magnetogram

36. ISSI Jan 2009 TRACE

37. ISSI Jan 2009 TRACE on MDI

38. ISSI Jan 2009 Hinode XRT

39. ISSI Jan 2009

40. XRT on MDI

41. ISSI Jan 2009 XRT on MDI

42. ISSI Jan 2009 XRT on MDI

43. ISSI Jan 2009 SOT FG V magnetogram

44. ISSI Jan 2009 UT Time on 1-2 Mar 2007 Magnetic Flux in box ( ) 3.0 3.4 1.0 19:0000:0005:00 (Cf. flux of whole region: ~ Mx)

45. ISSI Jan 2009

46. This initial observation from Hinode of a filament eruption supports the idea that flux changes in or near the eruption site are responsible for pre- explosive phase (e.g., slow-rise phase) dynamics. Flux changes = flux emergence and/or flux cancelation (also: tether weakening, slow tether cutting). (E.g., van Ballegooijen & Martens 1989, Moore & Roumeliotis1992, Rust & Kumar 1996, Lin & Forbes 2000, Chen & Shibata 2000, Feynman & Ruzmaikin 2004, Williams et al. 2005, Sterling, Harra & Moore 2007, Mikic et al. 2007.) Discussion

47. ISSI Jan 2009 Conclusions from this Hiinode Event Supports that pre-eruption (pre-flare) filament slow-rise phase is due to early flux changes (in this case: cancelation; slowly- driven tether-cutting reconnection). During pre-eruption period, = ~ 5% of flux of total erupting system. Therefore, the cancelation triggers release of the energy contained in the sheared field; it does not power the eruption. A question: What triggers the fast-rise phase? Hard to determine; see, e.g., Moore & Sterling 2006, Chifor et al. 2007. (Also, Liu et al. 2008, ApJ; suggest kinking plus internal tether cutting.)

48. ISSI Jan 2009 Quiet-region filament eruption of 28 Feb 2001 (Sterling, Harra, & Moore 2007; same event as Marque et al. 2002)

49. ISSI Jan 2009

51. Sterling, Harra, & Moore (2007) EITSXT

52. ISSI Jan 2009

53. Cf. Feynman & Ruzmaikin (2004) (also, e.g., Li et al. 2006, and many others).

54. ISSI Jan 2009 Sterling, Harra, Moore (2007) Slow-rise phase: Tether-Weakening Reconnection?

55. ISSI Jan 2009 Some Questons, and Objectives How common is the slow-rise phase? What triggers the fast-rise phase? (TC, breakout, instability, something else?) What triggers the slow-rise phase? I suspect B cancelation and/or emergence. Examination of several more good events needed. More broadly: –Larger-scale consequences of slow rise phase (e.g., hints for breakout?). –Dimmings and remote connections (dittio). Need: –More good e.g.s (AR or QS). –B data.

56. ISSI Jan 2009 Cavity Connections More examples are needed, of course! If there is a particularly good e.g., then it may be ripe for modeling. Rise of cavity with time: Over the long term, could be easier to see than with filaments alone. (Gibson et al. 2006; S. Martin, some talks and priv. com.) That is, could help understand the long-term evolution toward eruption. Observations of cavities in (probably) EUV with SXRs could help understand the role (if any) of tether cutting at time of eruption. Chewy N? Comparing with B evolution may help understand long-term evolution. (van Ballegooijen et al.?) This may be difficult in practice, however: B measured on disk while cavity height measured at limb; cavity-related fields may be too weak to measure changes do to cancelation, etc., accurately.

57. ISSI Jan 2009 Sterling & Moore (2003)

58. ISSI Jan 2009

59. Sterling (2004)

60. ISSI Jan 2009 Sterling & Moore (2003)

61. ISSI Jan 2009

62. 01 August 2008 Eclipse from Jinta, Gansu, China

63. ISSI Jan 2009