Download From eclipse drawings to the coronagraph and spectroscopy

From eclipse drawings to the
coronagraph and spectroscopy
1
Hardi Peter
Kiepenheuer-Institut
für Sonnenphysik
Freiburg
solar eclipse, 11.8.1999, Wendy Carlos and John Kern
What we might see on Wednesday…
2
for before
Antalyathe– eclipse:
Linker, Mikic
et al.
Theeclipse
coronaprediction
one rotation
2 March
2006
global MHD model based on observed magnetogram
N
N
HAO / MLSO
Lasco C2 / SOHO
1
3
Why the Sun?
¾ it is our Star
¾ nearest and only star where we might resolve the basic processes
on their characteristic temporal and spatial scales
¾ the Sun is our ultimate source of energy — influences the whole heliosphere
¾ solar activity can cause failure of space based and terrestrial technical systems
¾ the Sun influences the Earth’s climate
¾
¾
¾
¾
the Sun is a huge plasma laboratory
dynamics of a magnetized plasma
how do stars and galaxies produce magnetic fields?
basic physical questions: e.g. the (solved) neutrino problem
¾ the Sun’s past: how was the Sun formed and why do we have (9) planets?
¾ the Sun’s future: how will the habitable zone change in the future?
¾ change of paradigms:
so far:
classification (partly) according
to instrumental capabilities
visible
UV
EUV/X-ray
particles
—
—
—
—
photosphere
chromosphere
corona
heliosphere
¬ but: strong coupling between various atmospheric structures !
Challenge: close interaction between modeling and observations:
models are needed for interpretation of increasingly complex models
... and the Sun (and its corona) is fascinating
4
fine loops
in the corona
amateur
photography
Eclipse drawing, 1860
the solar corona
modern coronagraph
2
Outline
5
1. From eclipse drawings to the coronagraph and spectroscopy
2. The solar atmosphere and magnetic field
3. Modern observational techniques
4. Coronal heating and energetics
5. Closed magnetic structures – loops
6. Open magnetic structures – coronal holes and the solar wind
7. Stellar coronae
8. The microstate of the solar corona and the solar wind
9. Space weather and solar–terrestrial relations
10. Structures, waves and turbulence in the heliosphere
6
the solar chromosphere, just before totality
3
First eclipse photo: Berkowski, Königsberg, 28.7.1851
7
daguerrotype
Real adventures … 1898 in India
8
4
Fundamental Physics: eclipse 1919
Drawing vs. photography
9
10
18. July 1860
Spain,
drawing following the eclipse,
Warren de la Rue
Desierto, Spain,
40 s exposure time,
Angelo Secchi
5
Solar eclipse 18.7.1860: more drawings
G. Tempel
F. A. Oom
von Feilitzsch
E.W. Murray
F.Galton
C. von Wallenberg
11
from: C.A. Ranyard (1879), Mem. Roy. Astron. Soc. 41, 520, Kap. 44.
12
… astrophysics … –– spectroscopy
Spectrum
screen
prism
(today: reflective grating)
slit
normal lamp:
solid state
(Wolfram-spiral)
black body continuum
hot solid body
(„black body radiator“)
––
continuous
spectrum
6
13
… astrophysics … –– spectroscopy
spectrum
screen
prism
slit
street lamp
at pedestrian crossing:
Sodium-lamp
hot solid body
(„black body radiator“)
––
continuous
spectrum
„emission line spectrum“
radiating gas
––
emission lines
14
… astrophysics … –– spectroscopy
spectrum
screen
prism
slit
Na
gas in front of
continuous radiator
hot solid body
(„black body radiator“)
––
continuous
spectrum
„absorption lines"
radiating gas
––
emission lines
gas absorbing radiation ––
absorption lines
7
15
Fingerprints of the elements
solar spectrum
Fraunhofer lines
KH
G
Ca
[Fe]
F
Hβ
E
D
Fe
Na
BA
C
Hα
terrestrial
spectra of gases
H
Na
Fe
Ca
Mg
16
A new element: Helium
absorption lines in the photosphere
Ca II
KH
[Fe]
G
Hβ Fe I
F E
solar eclipse 1868:
Janssen and Lockyer identify the up to then unknown „D3 line“
Æ new element: Helium
Hα
C
Na I
D
1
2
3
discovered in Earth's atmosphere not before 1895 (!)
emission line spectrum of the chromosphere
during a solar eclipse
8
17
Emission line spectrum of the chromosphere
¾ slit perpendicular to limb
¾ take spectrum just when moon covers photosphere but NOT chromosphere
’ only a short instance ’ flash spectrum
chromosphere:
flash spectrum
wavelength
Ca II K
Ca II H
"normal"
photospheric
spectrum
with absorption lines
18
Even more new elements ?
’
emission lines in planetary nebula
Nebulium ?? …
green line: 530.3 nm
yellow line: 569.4 nm
red line:
637.4 nm
⇔
⇔
⇔
Fe13+
Ca14+
Fe9+
first attributed to „Coronium“
Chemistry:
~1900: periodic system of the elements:
no room for "normal" new element !
SOHO / Lasco C1 / green line
more than 100 „corona lines“ identified in the visible:
The strongest ones:
atomic physics: systematic investigation of spectra
of atoms and ions
astrophysics: systematic comparison
to results of atomic physics
Bengt Edlén
(Lund)
&
Walter Grotrian (Potsdam)
(1933 – 1942) Æ lines from highly ionized elements
9
The corona is hot !!!
temperature of the corona:
– ca 106 K
– more than 100 x hotter than surface
19
~ 1 000 000 K
energy required to heat corona:
flux ~ 100 W/m² at photosphere
’ no problem!
(less than 10-6 of luminosity)
Problem:
How does a cold body
heats a warmer body ??
~6000
Grad
~ 5777
K
(surface)
(corona)
contradiction to laws of thermodynamics?
’ energy transport e.g. through waves
e.g. microwave oven
in the solar corona:
magnetic field is the "energy agent"
’ magneto-acoustic waves
’ induced currents
Prominences: cool structures in the hot corona
20
Prominence observed on
Skylab, 1973, He II (304 Å)
also visible during eclipses…
Grande Daddy prominence
4. June 1946, HAO, Hα
amateur shot
Franky Dubois
2.9.1999
10
21
Prominences and filaments
on the disk:
filament.
absorption of
light from the
photosphere
full Sun in Hα
(Meudon)
above the limb: prominence.
„grande curtain“
Prominences through spectroscopy outside eclipses
22
1860s observations:
– use prime focus of single lens (stray light)
– f ≈ 2 m ’ diameter of solar image: ~20 mm
procedure to observe above the limb
(following Secchi/Schellen, 1872)
1.
place slit parallel to limb
2.
now slowly increase width of slit
up to ¼ to ½ mm
3.
take care that the slit
"is not touching the disk"
≈ ¼ mm
this works only for strong isolated lines
e.g. Hα ("slitless spectrograph")
today: narrow band filtergraphs:
– visible: Lyot filter, Fabry-Perot and Michelson interferometers
– VUV:
FPI (?)
– EUV-X-ray: foils, multi-layer coating
11
Drawings of structures above the limb
23
using spectroscopic methods
outside eclipses
by Etienne Trouvelot (1827-95)
’ he was one of the
most skilled
scientific artists
of the pre-photographic era
The magnetic field structuring the corona
1.
2.
3.
25
magnetic field map of the photosphere (“solar surface”) J Zeeman effect
potential field extrapolation (or better)
compare to structures in the corona
solar eclipse, 30.June 1973, photograph by Serge Koutchmy
potential field extrapolation: Altschuler at al. (1977) Solar Physics 51, 345
12
26
The activity cycle of the Sun
the Sun in white light
maximum
29.5.1996
28.3.2001
Big Bear Solar Observatory
minimum
11 year cycle of the Sun:
Sunspot number
monthly
smoothed
¾ sunspot number
¾ magnetic polarity
¾ magnetic activity
(since 1843)
(since 1908)
basic mechanism:
D dynamo generating magnetic field
Year
The corona: maximum vs. minimum
Minimum
Maximum
¾
¾
¾
¾
¾
¾
¾
¾
“simple” dipolar structure
few active regions (sunspots)
prominent coronal holes
“helmet streamer” only at equator
18. 3. 1988, Philippines
27
complex magnetic structure
many active regions
almost no coronal holes
“helmet streamer” at all latitudes
High Altitude Observatory - NCAR
16. 2. 1980, India
13
28
The X-ray corona in the solar cycle
1993
1995
minimum
100 x brighter
1991
maximum
Yohkoh Soft X-ray Telescope (SXT), X-ray emission at about 1 nm
The corona outside eclipses: Lyot's coronagraph
29
essentials of the coronagraph: get stray light down to < 10-7 – 10-8 !!
occulting surface: light from solar disk reflected out of window
Lyot stop:
Lyot dot:
blocks light diffracted at entrance pupil / edges of objective
stops light from secondary reflections within objective
Lyot (1932), Z. Astrophys. 5, 73
exit windows
elongated
tube to
protect
lens from
dust
objective
– single lens for
minimal number of
scattering surfaces
– easy to take out
for frequent cleaning
achromatic
field lens:
images objective objective:
images
onto Lyot stop
corona
occulting
surface
image plane
of the Sun
Diaphragm Lyot dot
Lyot stop /screen
image plane
of objective /
entrance pupil
image
of
corona
14
What is seen during an eclipse?
limb
¾ (K) continuum corona
1
intensity relative to disk center
30
- no absorption lines
solar
disk
- polarised:
free electron scattering
¾ Fraunhofer corona
“normal” sky
clear blue sky
-6
10
-8
F-corona
¾ Line corona
total eclipse
L-corona
1
- not polarized:
dusk scattering
– Zodiac light…
K-corona
10
- absorption lines visible
5
distance from disk center
in solar radii
- emission lines:
e.g.: “green coronal line”
- emission of atoms / ions:
new elements?
helium, coronium
The Zodiac light
31
15
32
The corona is hot!
¾ intensity scale height: 0.1 R
(around 1900)
¾ K-corona: free electron scattering:
thermal speed of electrons:
most narrow spectral features: 6 nm
(Waldmeier 1941)
6 nm @ 500 nm ⇔ 4000 km/s ⇔ 600·103 K
(electron temperature)
¾ Emission lines of highly ionised species (Edlén & Grotian 1939-41)
green line: Fe XIV (530.3 nm)
yellow line: Ca XV (569.4 nm)
red line:
Fe X
J
(637.4 nm)
these ions exist only at > 106 K
¾ L-corona: line width of emission lines: green line: 0.08 nm
0.08 nm @ 530 nm ⇔ 45 km/s ⇔ 4·106 K
(ion temperature)
33
A static heat conduction corona: temperature
heating at the “base”
of the corona:
T
TC
below base:
R < r < rC
R rC
¾ equilibrium of
heat conduction and heating:
"heated aluminum pipe"
conductive
heat flux:
BC:
T ( r = R ) << TC
Integration in: R → rC
height r
⇒
following Unsöld (~1960)
typically: f0 = 100
W/m2
16
Why 106 K ? – a coronal thermostat
34
¾ thermal conductivity:
fW ∝ T 5/2
¾ more heating J T-increase
J more efficient heat conduction
TC ∝ f 0 2 / 7
J only small net T-increase
¾ same for less heating…
changing the
heating rate f0
by orders of magnitude
results only in a small
change of the
coronal temperature
f0 [ W/m2 ]
17600
370
0.29
TC [106 K]
5.0
3.0
0.5
I “solar like”
(Leer 1998)
¾ solar wind
¾ magnetically open regions: 90% of the energy input powers solar wind
¾ more heating J even more losses due to wind
J less energy to heat corona
35
Pressure of a static heat conduction corona
heat flux in a fully ionized plasma (Spitzer 1956):
idea: carry away energy to Sun and infinity
by heat conduction (spherical coordinates):
T
TC
R rC
r
solve hydrostatic equilibrium in outer part:
pressure at infinity
does not vanish !
17
36
Can a static corona exist ?
Compare pressure of static corona to interstellar medium (ISM)
ISM:
TISM ≈ 100 K
nISM ≈ 10 cm-3
-3
p* ≈ 1000 K cm
=Γ
The Sun: TC ≈ 106 K
nC ≈ 108 cm-3
rC ≈ 2 R¯
p
pC
’ pC ≈ 1014 K cm-3
’ Γ ≈ 10-5
pISM
pISM
p∞
’ p*∞ ≈ 109 K cm-3 À p*ISM
Corona has to expand!
’ coronal wind
’ thermally driven
’ pressure (gradient) driven
rC
Rheliosphere
r
However: Stars with cool outer atmospheres
TC ≈ 104 K ’ p∞ < pISM
’ no wind !!
Continuous corpuscular radiation: solar wind
37
tails of comets:
α
two-part structure:
– dust tail:
controlled through
radiation pressure
– ion tail
(→ polarized light)
interaction of
ionized particles
in cometary tail
with solar wind
com
et
Æ solar wind
comet Hale-Bopp
(Biermann ~1941)
angle α:
’ wind speed
~ 1000 km/s
solar wind at Earth:
Sun
velocity:
400 – 800 km/s
density:
1 – 10 protons / cm3
temperature: 100 000 K
18
38
Observing the solar atmosphere
photosphere
transition region
chromosphere
corona
1024
106
radio: 70 – 900 GHz
temperature [K]
1022
EUVlines
105
104
Hα wing
“normal”
ground-based
observations
Hα core
Ca II K1
K2
(600-1600 Å)
K3
1020
1018
He IR
particle density [m-3]
X-rays
1018
EUV-Continua (>912 Å)
100
1000
height above τ5000=1 [km]
10000
From the upper chromosphere to the hot corona
39
full Sun scans
of CDS / SOHO
19
Summary / lessons learnt
40
~ 1850: first systematic "modern" eclipse observations
~ 1870: introduction of spectroscopy into coronal physics
~ 1930: invention of coronagraph
~ 1940: coronal lines are from highly ionized species ’ the corona ~106 K
~ 1970: first advanced X-ray observations
¾ the corona is magnetically structured
¾ the appearance of the corona changes with solar activity cycle
¾ 106 K is "quite natural": heat conduction acts as thermostat
¾ a static hot corona cannot exist ’ expansion
¾ appearance the solar atmosphere changes dramatically with temperature
From eclipse drawings to the
coronagraph and spectroscopy
20