Early Chromospheric Response During a Solar Microflare Observed with RHESSI and SOHO ’s CDS

1. Early Chromospheric Response During a Solar Microflare Observed with RHESSI and SOHO ’s CDS Jeffrey W. Brosius (Catholic U at NASA’s GSFC) Gordon D. Holman (NASA’s GSFC)

2. Overview We use SOHO/CDS spatially resolved high time resolution (7.2 s) EUV spectra that cover a wide temperature range (3×10⁴ – 8×10⁶ K) to investigate a GOES B1.8 microflare. Result: O V emission enhancements precede those of Fe XIX & 3-12 keV, and O V has v ~ -200 km/s component for 3 minutes, with opposing velocities for about 1 minute.

3. Terminology Chromospheric evaporation: upward motion into the (low density) corona due to expanding chromospheric material heated by nonthernal particle beams and/or thermal conduction. Large energy fluxes produce “explosive” evaporation, which features upward motions in flare (~10 MK) plasma and downward motions in cool (~0.1 MK) plasma. Small energy fluxes produce “gentle” evaporation, which features only upward motions (Fisher etal. 1985).

4. History • Neupert (1968) first described chromospheric evaporation based on OSO-3 and radio observations. • BCS on both SMM and Yohkoh measured upflows ~300 km/s in 10 MK lines during flares; no spatial resolution (Antonucci etal. 1982, Mariska etal. 1993). • SOHO/CDS and Hinode/EIS raster images yield relative Doppler velocities with time resolution 3 min < cadence < 11 min (Czaykowska etal. 1999; Teriaca etal. 2003; Milligan etal. 2006, 2008; Raftery etal. 2009). • CDS rapid cadence (~10 s) stare spectra yield dynamics of flares (Brosius 2003, 2009; Brosius & Phillips 2004; Veronig etal. 2010), a flare-like transient (Brosius & Holman 2007), and microflares (Brosius & Holman 2009, 2010).

5. Obtaining Rapid Cadence (7.2 s) EUV Spectra with SOHO’s CDS: the FLAREDOP Study • Short exposures (1.5 s). • Spatial compression (six 4″×20″ slit pixels). • Limited spectral coverage: 516-526, 580-611, 623-631 Å. • Sit & stare; get images elsewhere. • Net time resolution is readout-limited to 7.2 s.

6. Previous Relative Timing Results • During GOES M2.3 flare of 2001 April 24, chromospheric evaporation was “gentle” during precursors and “explosive” during impulsive phase. The Fe XIX intensity rose and stayed above its preflare noise level during 2nd O V precursor, and the Fe XIX intensity began its impulsive rise nearly 90 s later than that of the cooler lines (Brosius & Phillips 2004, ApJ613, 580). • In FLT during GOES M1.6 flare of 2004 July 22, a slow rise of Fe XIX emission preceded the O V impulsive rise by 2 m, suggesting direct coronal heating. Chromospheric evaporation during impulsive phase was “explosive” (Brosius & Holman 2007, ApJ659, L73). • Brosius & Holman (2009, ApJ692, 492) observed gentle evaporation during a GOES B2 microflare on 2005 November 16. The microflare’s impulsive chromospheric and TR emission began 2.5 m before its coronal emission, and 2.8 m before its Fe XIX emission appeared above the noise. • Brosius (2009, ApJ701, 1209) observed conversion from explosive to gentle evaporation between successive Fe XIX bursts in an M1.5 flare on 2004 July 27. Initial slow increase of Fe XIX emission may have been due to direct coronal heating; first O V impulsive rise preceded that of Fe XIX by 1 min.

7. EIT 195 Å and RHESSI3-12 keV Images on 2009 July 5

8. CDS, RHESSI, GOES Light Curves

9. CDS O V 629.7 Å Sample Line Profile Showing Blue Component

10. CDS Line Intensity Ratios of Blue to Main Components

11. Timings Among Emission and Velocity Features in “A”

12. Timings Among Emission and Velocity Features in “B”

13. Summary • Microflare’s onset observed in TR (O V) and upper chromospheric (He I) emission 1 min before Fe XIX and 4 min before 3-12 keV hot plasma flare emission. • In later brightenings O V and He I increased 49 sec ≤ Δt < 2 min before hot plasma flare emissions, which likely indicates heating by nonthermal particle beam. • For 3 min during several bursts, O V and He I profiles had secondary, blueshifted (~ -200 km/s) components; for ~ 1 min the velocities of the blueshifted and main components were opposite, suggesting explosive evaporation. • No blueshifts in Fe XIX emission.

15. Abstract We observed a GOES B1.8 microflare in the Sun’s lone AR 11024 on 2009 July 5 with RHESSI and SOHO’s CDS. CDS provided rapid cadence (7.2 s) stare spectra within a narrow FOV toward the region’s center. Emission lines cover a wide range of temperature, and include He I (<0.025 MK), O V (0.25 MK), Si XII (2 MK), Fe XIX (8 MK). The start of initial burst in He I and O V line emission preceded the emergence of Fe XIX line emission by about 1 minute, and the emergence of 3-12 keV X-ray emission by about 4 minutes. Thus the onset of the flare was observed in upper chromospheric (He I) and transition region (O V) line emission before it was detected in high temperature flare plasma emission. This likely indicates the presence of a nonthermal particle beam early during the microflare. Similarly, in subsequent bursts the O V and He I intensities increase before emissions from the hot flare plasma. In intervals lasting up to about 3 minutes during several bursts, the He I and O V emission line profiles show secondary, highly blueshifted (~ -200 km/s) components; during intervals lasting nearly 1 minute the velocities of the primary and secondary components were oppositely directed, suggesting explosive chromospheric evaporation. This work was supported by NASA through SR&T grant NNX07AI09G.

16. Caption to Figure 1 (left) 4'×4' EIT 195 Å image obtained at 16:35:17 UT on 2009 July 5. The position of the CDS slit is outlined as six 4″×20″ segments into which its spatially resolved spectra were binned, centered around its commanded pointing at (+119″.8, -458″.2). The ≈10″ uncertainty in the CDS pointing is overplotted as a dotted rectangle. (right) 1'×1' RHESSI 3-12 keV image averaged over the 1-min time interval 16:36-16:37 UT. The CDS slit (and uncertainty) is again overplotted. RHESSI contour levels correspond to 10, 30, 50, and 70% of the maximum count rate in the RHESSI image cube. The microflare was observed in slit pixels “A” and “B.”

17. Caption to Figure 2 EUV light curves from CDS in “A,” along with GOES 1-8 Å and RHESSI 3-12 keV light curves. Fe XIX is smoothed with a 9-point running boxcar, and plotted on an absolute intensity scale; all other light curves are on arbitrary scales. Solid vertical lines indicate a relatively quiescent interval (15:20-15:40 UT) from which “reference” quantities (including wavelengths common to “A” & “B”) are derived. Note that the B1.4 and B1.8 microflares appear blended together when viewed in Si XII line emission.

18. Caption to Figure 3 Sample profile of O V 629.7 Å (spectral intensity in ergs cmˉ² sˉ¹ srˉ¹ Åˉ¹ vs. wavelength in Å) in “A” at 16:38:45 UT during the rapid rise of the secondary blueshifted component. The observed profile is displayed as asterisks, the “main” and “blueshifted” components are displayed as solid white and blue curves, and the net fitted profile is a solid red curve. Dotted vertical lines indicate ±3σ from the reference wavelength. Short solid vertical lines indicate the centroid wavelengths of the main and blueshifted components, for which the corresponding velocities are given in the upper left.

19. Caption to Figure 4 Intensity ratio of the secondary, blueshifted component to the main component for O V, He I, and Si XII in “A”. Horizontal lines indicate 2σ above the average ratios derived in the “reference” interval (15:20-15:40 UT) during which the secondary component is considered to be noise. Time intervals during which we find significant, secondary, blueshifted O V components are 16:35:40-16:36:30 and 16:38:14-16:41:27 UT in “A” and 16:35:40-16:37:31 and 16:38:14-16:40:03 UT in “B”.

20. Caption to Figure 5 Normalized light curves (left) and relative Doppler velocities (right) in “A”. For intervals in which two components are observed, the displayed intensity is the sum of the primary and secondary components, and the velocity of the blueshifted component is divided by 5 to reduce the vertical plot range. Start times for various events are indicated with solid vertical lines, including (a) the start of the O V precursor burst at 16:26:02 UT, (b) start of the Fe XIX intensity increase at 16:26:46, (c) start of the 3-12 keV emission at 16:30:10, (d) start of the rapid O V intensity increase at 16:34:37, (f) Fe XIX between-burst minimum at 16:35:26, (g) start of a later rapid O V intensity increase at 16:38:14, (h) Fe XIX rapid intensity rise to the main phase plateau at 16:40:10 UT.

21. Caption to Figure 6 Normalized light curves (left) and relative Doppler velocities (right) in “B”. For intervals in which two components are observed, the displayed intensity is the sum of the primary and secondary components, and the velocity of the blueshifted component is divided by 5 to reduce the vertical plot range. Start times for various events are indicated with solid vertical lines, including (m) the start of the O V precursor burst at 16:26:09 UT, (n) start of the Fe XIX intensity increase at 16:29:14, (c) start of the 3-12 keV emission at 16:30:10, (d) start of the rapid O V intensity increase at 16:34:37, (o) Fe XIX between-burst minimum at 16:36:15, (g) start of a later rapid O V intensity increase at 16:38:14, (p) Fe XIX rapid intensity rise to the main phase plateau at 16:39:28 UT. For 16:36:45-16:37:31 and 16:39:08-16:40:03 UT the main O V component is redshifted while the secondary component is blueshifted, suggesting explosive evaporation.

22. Microflare of 2005 November 16 • CDS light curves of He I (/1.5), Si XII (×7), O V, Fe XIX (×10). • Two precursors before microflare, with enhancement factors 1.3, 1.7 (He I), & 2.4, 3.8 (O V). • Chromospheric & transition region emission increase first, indicating nonthermal electrons. • 20 s impulsive EUV burst during which Fe XIX starts increase. • Burst enhancement factors: 3.2 for He I, 11 for O V, 2.0 for Si XII; 24 for O III, 2.4 for 1600 Å. • Fe XIX (and GOES 1-8 Å) emission lasts only about 2 minutes. • Gentle chromospheric evaporation, with upward velocities about -20 km/s.

23. TRACE 195Å Images With Part of CDS Slit in 27 July 2004 M1.5 Flare

25. Fig. 1: MDI Magnetogram andEIT 195 Å Image

26. Caption to Figure 1 4'×4' (a) MDI magnetogram and (b) EIT 195 Å image obtained at the times indicated in each frame on 2009 July 5. The position of the CDS slit is outlined as six 4″×20″ segments into which its spatially resolved spectra were binned, centered around its commanded pointing at (+119″.8, -458″.2). The ≈10″ uncertainty in the CDS pointing is overplotted as a dotted rectangle. The microflare was observed in slit pixels “A” and “B,” indicated in (b).

27. Figure 2: CDS, RHESSI, GOES Light Curves

28. Caption to Figure 2 EUV light curves from CDS in “A” and “B”, along with GOES 1-8 Å and RHESSI 3-12 keV light curves. Fe XIX is smoothed with a 9-point running boxcar, and plotted on an absolute intensity scale; all other light curves are on arbitrary scales. The solid vertical lines near the beginning of both frames indicate a relatively quiescent interval (15:20-15:40 UT) from which “reference” quantities (including wavelengths common to both frames) are derived.

29. Figure3: RHESSI Images of Two Microflare Brightenings

30. Caption to Figure 3 RHESSI 3-12 keV images showing two bursts during the microflare. The 1-min time interval for each image is indicated in the upper right. “Before” frames are on the left, and “during” on the right. The CDS slit (and uncertainty) is overplotted as in Fig. 1. Contour levels are the same in all frames, and correspond to 10, 30, 50, and 70% of the maximum count rate in the RHESSI image cube. All images are displayed on the same brigtness scale.

31. Fig. 4: CDS O V 629.7 Å Line Profiles Showing Rise of Blue Component

32. Caption to Figure 4 Sequence of O V 629.7 Å profiles (spectral intensity in ergs cmˉ² sˉ¹ srˉ¹ Åˉ¹ vs. wavelength in Å) showing the rapid rise of the secondary blueshifted component from 16:38:14 to 16:39:00 UT in “A”. The central time of each exposure is indicated in the upper left. The observed profile is displayed as asterisks, the “main” and “blueshifted” components are displayed as solid white and blue curves, respectively, and the net fitted profile is a solid red curve. Dotted vertical lines indicate ±3σ from the reference wavelength. Short solid vertical lines indicate the centroid wavelengths of the main and blueshifted components, for which the corresponding velocities are given in the upper left of each frame.

33. Figure 5: CDS Line Intensity Ratios of Blue to Main Components

34. Caption to Figure 5 Intensity ratio of the secondary, blueshifted component to the main component for O V, He I, and Si XII in “A” and “B”. Horizontal lines indicate 2σ above the average ratios derived in the “reference” interval (15:20-15:40 UT) during which the secondary component is considered to be noise. Time intervals during which we find significant, secondary, blueshifted O V components are 16:35:40-16:36:30 and 16:38:14-16:41:27 UT in “A” and 16:35:40-16:37:31 and 16:38:14-16:40:03 UT in “B”.

35. Figure 6: Timings Among Emission and Velocity Features in “A”

36. Caption to Figure 6 Normalized light curves (left) and relative Doppler velocities (right) in “A”. For intervals in which two components are observed, the displayed intensity is the sum of the primary and secondary components, and the velocity of the blueshifted component is divided by 5 to reduce the vertical plot range. Start times for various events are indicated with solid vertical lines, including (a) the start of the O V precursor burst at 16:26:02 UT, (b) start of the Fe XIX intensity increase at 16:26:46, (c) start of the 3-12 keV emission at 16:30:10, (d) start of the rapid O V intensity increase at 16:34:37, (f) Fe XIX between-burst minimum at 16:35:26, (g) start of a later rapid O V intensity increase at 16:38:14, (h) Fe XIX rapid intensity rise to the main phase plateau at 16:40:10 UT.

37. Figure 7: Timings Among Emission and Velocity Features in “B”

38. Caption to Figure 7 Normalized light curves (left) and relative Doppler velocities (right) in “B”. For intervals in which two components are observed, the displayed intensity is the sum of the primary and secondary components, and the velocity of the blueshifted component is divided by 5 to reduce the vertical plot range. Start times for various events are indicated with solid vertical lines, including (m) the start of the O V precursor burst at 16:26:09 UT, (n) start of the Fe XIX intensity increase at 16:29:14, (c) start of the 3-12 keV emission at 16:30:10, (d) start of the rapid O V intensity increase at 16:34:37, (o) Fe XIX between-burst minimum at 16:36:15, (g) start of a later rapid O V intensity increase at 16:38:14, (p) Fe XIX rapid intensity rise to the main phase plateau at 16:39:28 UT. For 16:36:45-16:37:31 and 16:39:08-16:40:03 UT the main O V component is redshifted while the secondary component is blueshifted, suggesting explosive evaporation.

39. Summary • Microflare’s onset observed in TR (O V) and upper chromospheric (He I) emission 1 min before Fe XIX and 4 min before 3-12 keV hot plasma flare emission. • In later brightenings O V and He I increased 49 sec ≤ Δt < 2 min before hot plasma flare emissions, which likely indicates heating by nonthermal particle beam. • For 3 min during several bursts, O V and He I profiles had secondary, blueshifted (~ -200 km/s) components; for ~ 1 min the velocities of the blueshifted and main components were opposite, suggesting explosive evaporation. • No blueshifts in Fe XIX or Si XII emission.