Active Region Upflow Plasma and its Possible Contribution to the Slow Solar Wind J. L. Culhane 1, D.H. Brooks 2, L. van Driel-Gesztelyi 1,3,4, P. Démoulin 3, D. Baker 1, M. L. DeRosa 5, C. H. Mandrini 6,7, L. Zhao 8, T.H. Zurbuchen 8 1 University College London, Mullard Space Science Laboratory, Dorking, UK 2 Naval Research Laboratory, Washington DC, USA 3 Observatoire de Paris, LESIA, Meudon, France 4 Konkoly Observatory, Hungarian Academy of Sciences, Budapest, Hungary 5 Lockheed Martin Solar and Astrophysics Lab oratory, Palo Alto, CA, USA 6 Instituto de Astronomia y Fisica del Espacio (IAFE), CONICET-UBA, Buenos Aires, Argentina 7 Facultad de Ciencias Exactas y Naturales (FCEN), UBA, Buenos Aires, Argentina 8 D ept. of Atmospheric, Oceanic and Earth Sciences, Univ. of Michigan, Ann Arbor, MI, USA,

2 Hinode XRT / EIS Discovery of Persistent AR Upflows EIS observed upflows of ~ 50 km/s (Harra et al., 2008) in the region of upflow seen by XRT (Sakao et. al., 2007) - persistent upflows seen from AR peripheries with temperatures: 1 MK ≤ T e ≤ 2.5 MK - morphology differs from fan-loops where plasma is downflowing with T e ~ 0.6 MK - upflows mainly originate at sites of Quasi-Separatrix Layers (QSLs) Comment by Sakao et al that AR upflows could contribute ~ 25% to Slow Solar Wind has led to significant interest in the topic but they did not suggest any outflow paths

3 AR Plasma Outflow Paths (van Driel-Gesztelyi et al., 2012) AR1/AR2 observed by Hinode/EIS in Fe XII emission (a) with significant NE upflow at AR1 (b) - LFFF extrapolation (c) shows QSL footprints and long, possibly open, field lines from QSL and upflow region LFFF model shows null at 102 Mm height between AR1 and AR2; supported by potential field model (d) - these models only describe the AR1/AR2 neighbourhood Global PFSS model addresses full-Sun (a) and shows AR1 fully covered by streamer (b; yellow) Null point and its spine field line extend to PFSS source surface at 2.5 R ʘ - provides upflow plasma pathway to heliosphere - related AR-composition plasma later detected by ACE spacecraft AR 2 AR 1 NULL

4 Summary of Topics Covered AR passed disc centre 10 – 15 Dec. 2007; upflows studied by Hinode/EIS - plasma parameters ( T e, n e, v, FIP-bias ) measured as f(t ) Upflow inclinations to LoS estimated by Démoulin et al (2013) from systematic changes in the flows with solar rotation - linear force-free field (LFFF) extrapolation for local AR field gave consistent upflow results and allowed identification of Quasi-Separatrix Layers (QSLs) NSO/GONG PFSS model shows that the inward Heliospheric Current Sheet (HCS) projection bisects AR Global PFSS model shows AR completely covered by helmet streamer closed field - not clear how any upflowing plasma could reach heliosphere and contribute to slow solar wind Back-mapped ACE in-situ plasma data show that increases in O +7 /O +6, C +6 /C +5 and Fe/O ( FIP bias proxy ) are present from just ahead of the HCS crossing West of the AR - this looks very like AR-originating material Mandrini et al suggest a mechanism for upflow material from AR to reach the heliosphere following at least two reconnections and be detected by ACE

5 AR and Associated Upflows From 10 – 15 Dec. 2007, AR passed Central Meridian (CM) - AR and the two principal associated upflow regions at its boundaries are shown below Upflows persistent during 5 day interval - XRT images are with Ti/Poly filters; EIS velocity maps from Fe XII/ Å line profiles - apparent upflow velocity reversal from West to East due to flows’ changing inclination to line-of-sight 10 Dec. 2007, 00:07 UT 12 Dec. 2007, 11:43 UT 14 Dec. 23:59 UT 10 Dec. 2007, 00:19 UT 12 Dec. 2007, 11:43 UT 15 Dec. 2007, 00:13 UT

6 Upflowing Plasma Properties Plasma parameters for East and West upflows - temp. ( T e ), density ( n e ), FIP-bias ( f FIP ) n e estimated from Fe XIII line intensity ratio ( I Å /I Å ) T e and FIP-bias are obtained from Differential Emission Measure (DEM) analysis FIP-bias values 3.0 ≤ f FIP ≤ 4.0 characteristic of slow solar wind Démoulin et al., 2013 studied both upflows in detail - flows are spatially coherent thin fan-like structures - inclined to LoS with  East  ~ and  West  ~ velocity and temporal evolution similar for all lines - strong stationary component suggests days/weeks duration for driving mechanism Flows usually located at Quasi-Separatrix Layers (QSLs)

7 Linear Force Free Field (LFFF) Extrapolation for AR Topological analysis in the linear force-free field extrapolation (Démoulin et al., 2013) allows identification of Quasi-Separatrix Layers (QSLs: black lines) Upflows originate from these QSL sites Upflows from reconnection at QSLs between dense AR loops and long low-density loops Reconnection at QSLs results from AR growth and dispersion LFFF magnetic field configurations are valid close to the AR - computed field lines at AR borders have similar inclinations to those shown by direct modelling of the EIS upflows - they do not imply that the upflowing plasma leaves the Sun QSLs

8 AR related to Global Magnetic Field Configuration NSO/GONG PFSS model: open field reaches the source surface at 2.5 R ʘ - inward projection of HCS bisects AR and separates East and West upflow regions - red/green areas show –ve/+ve polarity of open B-field regions - grey area shows mainly closed B-field regions and LoS magnetogram features for AR HCS Carrington (CR 2064) display of a Stereo-B EUVI image shows AR and two nearby opposite polarity CHs

9 PFSS model for 12-Dec-07 shows large-scale topological structures and AR helmet streamer separatrix surface in semi-transparent yellow (a) - AR shown without helmet surface (b) - AR in closed–field region fully covered by streamer - streamer is bordered by the two adjacent CHs, E and W of the AR - continuous blue line (a) shows the HCS Global Solar Magnetic Field Configuration Right-hand panel (b) has the helmet surface removed for clarity - LoS magnetic field structures of AR are shown - spine field lines (light-blue) do not enter the open field domain but remain closed below the streamer - thus long low density loops carrying the upflowing plasma are fully contained below the streamer - not obvious how plasma with AR composition could gain access to the Heliosphere

10 ACE in-situ Observations Related to AR Passage Back-mapped ACE data shown relative to a) ST-B EUVI 195 Å synoptic map for CR 2064 Data include: b) v p c) He/p, d) O +7 /O +6 and C +6 /C +5, e ) Fe/O, f) B radia l, g) B abs - red line indicates change in B radial polarity seen at ACE which shows HCS crossing Increases in O +7 /O +6, C +6 /C +5 and Fe/O (FIP bias) are present from West of the HCS crossing - active region material in a slow solar wind flow - reduction in He/p indicates streamer tip contribution Fast solar wind parameters are evident before and after the HCS transition Fast wind flow originates from the two adjacent CHs that are E and W of AR a) b) c) d) e) f) g) 30 Nov, Dec, 2007

11 N1 Possible Mechanism for Plasma Escape from AR N1 AR loops (blue) reconnect with large scale network fields (red) at the Eastern QSL - reconnection driven by AR expansion (green arrows) puts AR upflow plasma into long low density network loops These large scale loops reconnect with the open field lines (pink) at N1 that are associated with the Northern CH Mandrini et al., (2014) analysed the global magnetic topology - located four high altitude magnetic null points within ± 20 0 latitude of AR only one of these (N1) has associated open field lines

12 AR Upflow Contribution to Slow Solar Wind View of AR from 45 0 longtitude shows open field lines bending towards the ecliptic while for 0 0 longtitude field lines seen directed towards the Sun-ACE line Open field lines bend towards the ecliptic and so can deliver the upflowing plasma from East of the AR to the Sun – ACE line before the HCS passage i.e. from the North- West N1 Longitude 0 0 Longitude 45 0 E

13 Conclusions Discovery of AR upflows and identification of path to Heliosphere were sumarized Detailed measurement of upflow plasma properties for AR in December, 2007 showed typical active region values which are also found in the slow solar wind PFSS models show: a) HCS bisects the AR, b) AR completely covered by helmet streamer ACE in-situ plasma data were displayed relative to a Carrington synoptic map (CR 2064) for the interval 30-NOV-2007 to 27-DEC-07 which includes the disc transit of AR ACE data backmapped and related to polarity change in B r - clear that speed and composition assume slow wind characteristics West of the AR - given streamer coverage of the AR on both sides of the HCS, not clear how AR plasma escapes Mandrini et al. show how upflow material originating from East of the AR can reach ACE from West of the AR following at least two magnetic field reconnections Further study needed to establish level of AR plasma contribution to Slow Solar Wind

14

15 Back-mapping Plasma Flow from ACE (L1) to the Sun Back-mapping from the ACE spacecraft is in two stages - radial ballistic mapping from L1 to the 2.5 R ʘ source surface - PFSS extrapolation from source surface to photosphere - uncertainty ~ 10 0 ACE proton speed data shown on ST-B EUVI 195Å image - proton speed (Sun to ACE) shown with red/fast and blue/slow solar wind - ACE solar footprint, HCS and B radial also indicated ACE Footprint HCS BrBr

16 Some of the plasma flows remain confined, others get out and from magnetic modeling we can say exactly which ones. The coronal hole provides a route for the upflows to become outflows. van Driel-Gesztelyi, Culhane et al., Solar Physics, Modeling where the slow wind comes from