First results of the Chromospheric LAyer

Spectro-Polarimeter (CLASP2)

Ryohko Ishikawa(1)

David McKenzie(2), Javier Trujillo Bueno(3), Frederic Auchere(4), Ryouhei Kano(1), Donguk Song(1), Masaki Yoshida(1), Toshihiro Tsuzuki(1), Fumihiro Uraguchi(1),

Takenori J. Okamoto(1), Laurel Rachmeler(2), Ken Kobayashi (2), and CLASP2 team

(1) National Astronomical Observatory of Japan, (2) NASA Marshall Space Flight Center, (3) Instituto de Astrofisica de Canarias, (4) Institut d'astrophysique spatiale

1

• The magnetic energy is significant; ~2x106 erg/cm2/sec (Ishikawa+2008)

2

heating and high-frequency waves by magnetic reconnection with pre-existing vertical fields

Implications from MHD simulation (Isobe+2008)

(420 km)

Heigh

t

Field Strength from Stokes inversionz

x1,320 km

x

yContinuum

(e.g., Martinez Gonzalez+2010, Gomory+2010)

Ishikawa+2010

Required energy (Withbroe & Noyes 1977)• QS corona: 3x105 erg/cm2/sec• QS chromosphere: 4x106 erg/cm2/sec

Photospheric Magnetic Fields Revealed by Hinode

Example: Granular-scale horizontal fields in the quiet Sun

Chromospheric Dynamics Revealed by Hinode

3

Ubiquitous jets

Shibata+07

De Pontieu+07Okamoto+07

MHD waves

prominence

spicule

solar disk

Many MHD fundamental processes

1000km

1000km

Growing Demand: Magnetic Field Measurement in Upper Solar Atmosphere

4

Quiet photosphere

(β>1)

Dynamic chromosphere

(β<1)

Hinode observation ©NAOJ

β=1

Hei

ght

[Mm

]

Corona

Photosphere

Gary (2001)

Chromosphere

Plasma β (Pgas/PB)

Present

Next

Magnetic field measurement

UV Spectral Lines: Access to the Base of Corona• Sensitive to physical properties at the layer where the temperature suddenly

increases (upper chromosphere and transition region)

5

• Zeeman diagnostics is limited in UVHI Lya (121 nm)

Mg II h & k (280 nm)

ν

Δ"# =%2m()

Stokes-V (circular polarization)

DwB

DwB

[Exception] Mg II h & k in magnetized regions

-

σ+

σ-

Height [km]

Tem

pera

ture

photospherechromospheretransition region

Scattering Polarization & Hanle Effect

6

Atom

Disk center obs.

NOpolarization

Off-limb obs.

Linear polarization parallel to limb

Linear pol.(Stokes-Q)

Parallel to the limb

Perpendicular to the limb

1D plane-parallel atmosphere

Scattering Polarization & Hanle Effect

7

AtomOff-limb obs.

Linear polarization non-parallel to limb

Linear pol.(Stokes-Q)

Parallel to the limb

Perpendicular to the limb

1D plane-parallel atmosphere

B

Magnetic field modifies the scattering polarization

*+,~"# =%2m().

Hanle effectHanle effect

k2V k2R

k1V k1R

k3

Zeeman effect

�

Theoretical calculation with 1D model atmosphere, Mg II k

Alsina Bellester+2016

B=0G

B=0G

10G 10G20G 20G

Q/I (perpendicular/parallel to the limb) U/I

9

8. . OTNDING OC ES 1VPEQIMENSR-/6. : &) /6. : &

CIENCE 9 ECSIUER IN ( RSEPR( ED MDH HA CD C% DLDH !. ( :L M H% HE D M P

) 1 M MDH HA L MM D HE D MDH D :L M E ED L1 M MDH HA MC 3 E S M

+ 2O EH MDH HA MC MD R E L DC H HL C M LDMDH DH

0 DM- : P CHMH <CDM L DLLDE

04 !/C H HL C D 6P %.E C M H%:HE D M - 4 5 6 M & & NM04 ) !/C H HL C D 6.P M H%:HE D M - G 55 H H NM

Final Goal: Establish a means to diagnose the magnetic field at the base of corona with UV spectro-polarimetry

CLASP CLASP2

10

/6. : CIENCE 4IGHLIGHSR :OLAQIMESQ

2IQRS 0ESECSION OF CASSEQING :OL IN

-200 -100 0 100 200X [arcsec]

600

700

800

900

1000

Y (s

olar

radi

us) [

arcs

ec]

-200 -100 0 100 200

600

700

800

900

1000

slit length: 400”

Stokes-Q Stokes-U

-0.10 -0.05 0.00 0.05 0.10

600

700

800

900

Y (s

olar

radi

us) [

arcs

ec]

-0.10 -0.05 0.00 0.05 0.10

600

700

800

900

-0.10 -0.05 0.00 0.05 0.10-0.10 -0.05 0.00 0.05 0.10 -0.10 -0.05 0.00 0.05 0.10-0.10 -0.05 0.00 0.05 0.10

[%]

-4

-2

0

2

4

-4

-2

0

2

4

wavelength [nm]

Clear center-to-limb variation up to 6% in Q/IFluctuating at a few% at ~10” both in Q/I and U/I

Q/I U/IIntensity

HI Lya wing (scattering pol. ONLY)

µ=0.2

Belluzzi+ (2012)

Kano+ (2017)

µ=0.4

Kano+ (2017)

11

2IQRS 0ESECSION OF CASSEQING :OL IN

-200 -100 0 100 200X [arcsec]

600

700

800

900

1000

Y (s

olar

radi

us) [

arcs

ec]

-200 -100 0 100 200

600

700

800

900

1000

slit length: 400”

Stokes-Q Stokes-U

wavelength [nm]

Q/I U/IIntensity

No clear center-to-limb variation (CLV) in Q/IFluctuating at a few of 0.1% both in Q/I and U/I

H I Lya core (scattering pol. & Hanle)

-0.10 -0.05 0.00 0.05 0.10

600

700

800

900

Y (s

olar

radi

us) [

arcs

ec]

-0.10 -0.05 0.00 0.05 0.10

600

700

800

900

-0.10 -0.05 0.00 0.05 0.10-0.10 -0.05 0.00 0.05 0.10 -0.10 -0.05 0.00 0.05 0.10-0.10 -0.05 0.00 0.05 0.10

[%]

-0.6

-0.4

-0.2

-0.0

0.2

0.4

0.6

-0.6

-0.4

-0.2

-0.0

0.2

0.4

0.6

Stepan+ (2015)

limb disk center

<B>=15G

<B>~250G

<B>=0G

CLV of spatial average of pol. with 3D MHD model

/6. : CIENCE 4IGHLIGHSR :OLAQIMESQ

Kano+ (2017)

Stepan+2018, Trujillo Bueno+2018

H H M D E H E ODMPMC MC H E DL D

12

9 REQUASIONAL 1UIDENCE OF 4ANLE 1 ECS• In Lyα core and Si III, U/I deviates from the positive and negative spatial

distribution due to local scattering as photospheric magnetic flux increases

CLASP/SJ

-100 -50 0 50 100X [arcsec]

600

700

800

900

Y (so

lar ra

dius)

[arcse

c]

-100 -50 0 50 100

600

700

800

900

DCAB

SDO/HMI

-100 -50 0 50 100X [arcsec]

-100 -50 0 50 100

R. Ishikawa et al. (2017)

0.0 0.5 1.0 1.5 2.0Magnetic Flux at Solar Surface (x10 ) [Mx]

0.0

0.2

0.4

0.6

0.8

1.0

Devia

tion f

rom Ex

pecte

d Sca

tterin

g Pola

rizati

on D

C

A

BLya wingSi IIILya core

18

Large

Zero

Lya coreHanle at B>10G*

Si IIIHanle at B>60G*

Lya wingNO Hanle

/6. : CIENCE 4IGHLIGHSR :OLAQIMESQ

121.50 121.55 121.60 121.65805

810

815

820

Y (so

lar ra

dius)

[arcse

c]

121.50 121.55 121.60 121.65805

810

815

820

Y (so

lar ra

dius)

[arcse

c]

+

-

+

-

U/I

wavelength [nm]

Q<0

dI/dx=dI/dy=0 Q=0

Q<0

Q>0

Q>0

U<0

U<0

U>0

U>0

U=0

IntensityStokes-Q Stokes-U

y (+Q)

x

μ=1

Q<0

dI/dx=dI/dy=0 Q=0

Q<0

Q>0

Q>0

U<0

U<0

U>0

U>0

U=0

IntensityStokes-Q Stokes-U

y (+Q)

x

μ=1

Stepan & Trujillo Bueno (2012), R. Ishikawa+(2017) *0.2BH

13

CLASP2 Instrument

CLASP CLASP2Observables Stokes-I, Q, U Stokes-I, Q, U, V

Spectral Lines Lya (121.6 nm) Mg II h & k at 280.0 nmResolutions 0.01nm (wavelength) & 2-3” (spatial) 0.01nm (wavelength) & 2” (spatial)Slit Length 400” 200”

Science Target Quiet Sun near the limb Quiet Sun near the limb & PlagePol. Precision 0.1% at 3σ

CCD

Primary mirror(Cold Mirror coat)

Magnifier(hyperbolicandfoldmirrors,NDfilter)

Cassegrain telescope

ApertureHeat absorber

Slitjaw Optics (SJ) Spectro-Polarimeter

(SP)

Slit mirrorrotating waveplate

-1st order

+1st order

off-axis parabola

CH1

CH2Spherical Grating

Polarizer

Tsuzuki+ in prep.

CLASP2 Development

14

Recovery of CLASP instrument

Vibration test of new structure @ JAXA

Integration & alignment in

NAOJ clean room

Packing of instruments

See poster B28 (Song et al.) about CLASP2 polarization calibration

See Song+ 2018, SPIE for SP alignmentand Yoshida+ 2018, SPIE for telescope alignment

CLASP2 Launch: April 11, 2019 16:51 UT @ WSMR

150 DM- : P CHMH <CDM L DLLDE

16

Observation Sequence:[1] 18 sec

Disk center for pol. cal.[2] 155 sec

Plage region[3] 134 sec

Quiet Sun near the limb

Flight Performance:Drift: 1”/minJitter: <±0.1” (P-V)

Coordinated observationHinode, IRIS, DST @ Sac Peak(BBSO)

17

/6. : :QELIMINAQ ERTLSR

0ESECSION OF CASSEQING :OL AQOTND G 55 H

[%]Temporally

averaged Q/ITemporally

averaged intensity3.0

1.5

0.0

-1.5

-3.0

Temporally & spatially averaged

I

Q/I

Belluzzi & Trujillo Bueno (2012)

• Wavelength structure of Q/I is consistent with the theoretical prediction (Belluzzi & Trujillo Bueno 2012).– Mg II k shows strong scattering polarization while Mg II h

shows exactly zero polarization at the line center.

slit: 200”

Theoretical calculation(µ=0.1, FAL-C, B=0G)

18

/6. : :QELIMINAQ ERTLSR

/ENSEQ SO LIM AQIASION OF 5• Clear CLV in far wings• No clear CLV is in the Mg II k center

and near wings

Pointing B (154s)

Pointing C (134s) Pointing A

Pointing BPointing C

Plage

Further investigation & discussion will be in Rachmeler+ in AGU

Plage

Plage

19

/6. : :QELIMINAQ ERTLSR

IGNI CANS 5 OUEQ SHE :LAGE

[%]Temporally

averaged V/ITemporally

averaged intensity2.0

1.0

0.0

-1.0

-2.0

Temporally & spatially averaged

I

V/I

• Induced by the longitudinal Zeeman effect• Many spectral lines (Mg II h & k, Mg II triplet, Mn I, etc….) show V/I

slit: 200”

Summary

• CLASP & CLASP2 Demonstrated that high-precision UV spectro-polarimetery is feasible & detected scattering polarization in UV for the first time

• CLASP2 provided detailed measurements of Mg II h & k polarization spectra– Linear polarization (Q/I & U/I) due to scattering– Circular polarization (V/I) due to Zeeman effect over several spectral lines: enable to

constrain magnetic field structures from lower to upper chromosphere in plage

• For a final goal– Utility of V/I in Mg II h & k for determining the magnetic field vector– Importance of simultaneous observations of Lya and Mg II h & k lines

20