Transcript An introduction to the MULTI radiative transfer code

An introduction to the MULTI
radiative transfer code
Lars Heggland
Institute of Theoretical Astrophysics,
University of Oslo
Kwasan Observatory, Kyoto University
Radiative transfer
The study of generation and transport of
radiation in stellar atmospheres
 Important physical processes:

Absorption
 Emission / reemission
 Scattering
 Formation of spectral lines

Why is this important?
Radiation is one of the few ways to
directly collect information about a star
 Spectral line formation is highly
dependent on physical parameters such
as:

Temperature (ionisation states, intensity)
 Velocity fields (Doppler shifts)

Why is this important?
Thus, a wealth of information is
contained in spectral lines
 By studying specific lines, we can get
information about conditions from the
photosphere (neutral or singly ionised
lines, 5-10000 K) to the corona (highly
ionised metal lines, 1 MK ++)

Numerical radiative transfer
The aim: using theoretical models to
explain and reproduce observations
 OR: using models to predict future
observations
 The problem is non-trivial due to the
complexity of the physical processes
and of the atmosphere; simplifications
required

Numerical radiative transfer

Real atoms have hundreds of different
energy levels
Very computationally intensive
 Many levels have little effect on the studied
line
 Make simplified, smaller atomic models


Compute one element at a time
The MULTI code
A non-LTE radiative transfer code
written by Mats Carlsson
 Written in standard compliant Fortran77; designed for portability
 Freely available for use:

http://www.astro.uio.no/~matsc/mul22/index.html
The MULTI code
Works in 1D; multi-dimensional analysis
can be done by computing several
different rays
 Uses a given background atmosphere;
dynamics (waves etc.) can be taken into
account by running a simulation for
each timestep

The MULTI code
Powerful, but complex
 The amount of input data and
parameters is high
 Patience required, experience very
useful
 Impressive amounts of output data
make it worth it

Documentation
Main: A computer program for solving
multi-level non-LTE radiative transfer
problems in moving or static
atmospheres (available on Carlsson’s
website, 46 pages plus appendices)
 Update: mul22.ps (included in
distribution)
 multi.help, variables.doc (included)

Input files
ATOM: Atomic
model
(example: 6level calcium)
 Complex, but
needs only be
done once

CA 2
* ABUND
AWGT
6.3304 40.08
*NK NLIN NCNT NFIX
6
5
0
5
*
E
G
0.00000
2.00000
'CA II 3P6 4S 2SE
'
13650.248
4.00000
'CA II 3P6 3D 2DE 3/2'
13710.900
6.00000
'CA II 3P6 3D 2DE 5/2'
25191.535
2.00000
'CA II 3P6 4P 2PO 1/2'
25414.465
4.00000
'CA II 3P6 4P 2PO 3/2'
95785.470
1.00000
'CA III GROUND TERM '
*
F
NQ QMAX Q0 IW
GA
4 1 3.3000E-01 40 300. 3. 0 1.48E08
5 1 6.6000E-01 40 300. 3. 0 1.50E08
4 2 4.4200E-02 40
75. .3 0 1.48E08
5 2 8.8300E-03 40
75. .3 0 1.50E08
5 3 5.3000E-02 40
75. .3 0 1.50E08
* J I P
A0
TRAD ITRAD
6 1 1 2.0363E-19 5915. 2
6 2 1 6.1484E-18 5755. 2
6 3 1 6.1484E-18 5755. 2
6 4 1 2.3823E-18 4925. 2
6 5 1 2.3823E-18 4925. 2
* COLL
CA2COL
1.71E-07
1.71E-07 4.27E-07
2.92E-07 1.31E-06 2.08E-07
2.92E-07 2.93E-07 1.29E-06 3.05E-07
1.45E-10 1.88E-10 1.88E-10 2.68E-10 2.68E-10
ION
2
2
2
2
2
3
GVW
1.62
1.61
2.04
2.01
2.01
GS
3.0E-06
3.0E-06
3.0E-06
3.0E-06
3.0E-06
Input files


ATMOS: Model atmosphere (example: truncated
VAL3C)
Specified on lg column mass, lg optical depth (500) or
geometrical depth scale
VAL3C
MASS SCALE
* LG G
4.44
* NDEP
52
*LG COLUMN MASS
TEMPERATURE
NE
V
-5.279262E+00
4.470000E+05
1.205000E+09
0.
-5.270430E+00
1.410000E+05
3.839000E+09
0.
-5.269783E+00
8.910000E+04
5.961000E+09
0.
-5.268492E+00
5.000000E+04
9.993000E+09
0.
-5.267285E+00
3.700000E+04
1.318000E+10
0.
* HYDROGEN POPULATIONS
*
NH(1)
NH(2)
NH(3)
NH(4)
2.3841E+03
7.9839E-04
2.0919E-04
2.3110E-04
5.3401E+04
1.8790E-02
7.4560E-03
8.1751E-03
2.4030E+05
7.5740E-02
2.9400E-02
3.1550E-02
2.7390E+06
6.7709E-01
1.7230E-01
1.7180E-01
1.3850E+07
3.2580E+00
4.7581E-01
4.3231E-01
3.6271E+07
9.2240E+00
8.6389E-01
7.2180E-01
VTURB
1.128000E+01
9.870000E+00
9.820000E+00
9.760000E+00
9.730000E+00
NH(5)
2.9470E-04
1.0430E-02
4.0101E-02
2.1430E-01
5.2440E-01
8.5328E-01
NP
1.0030E+09
3.1990E+09
5.0310E+09
9.0170E+09
1.1970E+10
1.3710E+10
Input files
DSCALE: Depth scale to use for
calculations
 Does not need to use the same values
as the atmosphere model; interpolation
is performed

MV45C3
MASS SCALE
45 -6.672232
-5.22498
-5.21206
-5.20977
-5.20795
-5.20536
-5.20078
-5.19238
Input files

INPUT: Run and output options
DIFF=5.0,ELIM1=0.01,ELIM2=0.001,QNORM=12.85,THIN=0.1,
IATOM2=0,ICONV=1,IHSE=0,ILAMBD=0,IOPAC=1,ISTART=2,ISUM=0,
ITMAX=300,ITRAN=0,NMU=3,
IWABND=0,IWATMS=0,IWATOM=0,IWCHAN=0,IWDAMP=0,IWEMAX=1,IWEQW=0,
IWEVEC=0,IWHEAD=0,IWHSE=0,IWLGMX=1,IWLINE=0,IWLTE=0,IWN=0,IWNIIT=0,
IWOPAC=0,IWRAD=0,IWRATE=0,IWSTRT=0,IWTAUQ=0,IWTEST=0,IWWMAT=0,
IWARN=2,IOPACL=0,ISCAT=0,INCRAD=0,INGACC=1,ICRSW=0,
IDL1=1,IDLNY=1,IDLCNT=1
Controls starting approximation, number
of iterations, convergence limit…
 Trial and error required for best results

Input files
ABUND, ABSDAT: used to calculate
background opacities
 Should not need to be changed in a
solar photosphere/chromosphere model
 …maybe in the corona and in other
stars

Output
Lots of data! (See variables.doc)
 Intensity, flux, line source functions,
population densities, transition rates…
 Data written to be readable by IDL;
reading and analysis routines are
included in the distribution

Sample output: C I line in
transition zone
(pdf file)