Radiative Transfer

Hélène Frisch

Radiative Transfer
An Introduction to Exact and Asymptotic
Methods
Hélène Frisch
Observatoire de la Côte d’Azur, CNRS,
Laboratoire Lagrange
Université Côte d’Azur
Nice Cedex 04, France

ISBN 978-3-030-95246-4 ISBN 978-3-030-95247-1 (eBook)

https://doi.org/10.1007/978-3-030-95247-1

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland
AG 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To Uriel, Anne, and Thomas
Preface

Transport of particles such as photons, neutrons, electrons, and molecules through

a host medium usually takes place by multiple (repeated) scatterings. This book
presents methods for constructing exact solutions of stationary radiative transfer
problems related to the transport of photons in a stellar or planetary atmosphere
or any foggy medium. When the scatterings do not modify the physical properties
of the host medium, the equations describing the transport of particles are linear.
In spite of this, neither the construction of exact solutions nor numerical solutions
are straightforward, because of the nonlocal character of multiple scattering. Exact
solutions provide trustful descriptions of the physical phenomenon and can serve as
testing ground for numerical methods of solutions.
Exact solutions, that is solutions which can be expressed as known functions
or quadratures, can be constructed for a few, but widely used, types of scattering
processes and geometries of the host medium. Scattering processes allowing exact
solutions discussed in this book are monochromatic scattering, complete frequency
redistribution, and Rayleigh scattering. Monochromatic scattering and Rayleigh
scattering (which accounts for the linear polarization and blue color of the sky)
govern the formation of continuous spectra. Photons change directions at each
scattering but keep the same frequency. Complete frequency redistribution (an
idiomatic expression of the astrophysical literature) governs the formation of
most atomic spectral lines. Photons do undergo changes of direction, but also of
frequencies, in an uncorrelated way, within the frequency range of the spectral line.
Concerning the host medium, the spreading of a source of light by a large number
of scatterings can be studied by assuming a medium of infinite extension. A standard
Fourier transformation can then provide the relevant solution. In astrophysics, and
also in nuclear reactors, a very important question is how a radiation field will
be reflected and transmitted through a scattering medium, such as a stellar or
planetary atmosphere. A closely related question is how photons created inside these
atmospheres will escape the medium, and what will be their observable distribution
in directions and frequencies. A reasonably good model for a stellar atmosphere is
a semi-infinite medium. Finding exact solutions is possible but rather hard. The first
exact solution of a transport problem in a semi-infinite plane-parallel medium was

vii
viii Preface

constructed by (Wiener and Hopf 1931) for the Milne problem, which describes the
temperature profile of a stellar atmosphere in radiative equilibrium. Their method
of solution, which now carries their name, is one the greatest achievements in
mathematical physics in the first half of this century. It relies on the properties of
analytic functions in the complex plane. In contrast, when the question is posed to
find the distribution of photons that have been reflected or transmitted through a
scattering medium of finite extension, a slab for example, then there is no explicit
expression for the radiation field, inside or outside the medium. However, there
are many interesting exact relations and integral equations, which can be solved
numerically. This topic is only briefly mentioned in this book.
Following the work of Wiener and Hopf, several methods have been developed
to tackle semi-infinite medium problems. One of the best known method in
astrophysics is the construction of a nonlinear integral equation for the famous H -
function, a sort of special function for half-space transport problems. Proposed for
the first time by Ambartsumian (1942), it largely avoids complex plane analysis and
relies on invariance properties of the radiative transfer process. Plasma physicists
became strongly interested in the subject in relation to the transport of neutrons
in nuclear reactors. New methods were developed, such as the Case singular
eigenfunction expansion method (Case 1960), which leads to linear singular integral
equations with Cauchy-type kernels. Methods of solutions for these equations,
also based on complex plane analysis, rely largely on a technique introduced by
Carlemann (1922) and developed by Muskhelishvili (1953). These methods lead
to boundary value problems in the complex plane, known as Riemann–Hilbert
problems, and provide explicit expressions for the H -function, as does the Wiener–
Hopf method.
A key motivation for writing this book was to present in the same volume several
of the methods leading to exact results for semi-infinite media, usually presented
in separate books or articles, for both scalar and polarized radiation fields, and also
connections, that are not always clearly apparent, between the various methods.
When the number of scatterings in the host medium is very large, asymptotic
techniques may be employed to analyze the large-scale behavior of the random
walk of the photons and to explain, for example, why monochromatic scattering has
all the characteristics of an ordinary diffusion process, while complete frequency
redistribution of spectral lines has in contrast those of a Lévy walk. This asymptotic
approach is presented in Part III of the book, while exact methods are presented in
Parts I and II. Part I deals with unpolarized radiation and Part II with monochro-
matic radiation, polarized by Rayleigh scattering, and spectral lines, polarized by
resonance scattering and the Hanle effect, which is a modification of resonance
polarization by a weak magnetic field. Details on the organization of each part are
given in Chap. 1.
The intended readership for the book ranges from first-year graduate students
to professional scientists. Astrophysicists can find in this book exact methods of
solutions used in radiative transfer, but also applied in distant fields, such as financial
mathematics. A major effort has been made in the organization of the material
to give a synthetic view on exact and asymptotic methods in radiative transfer.
Preface ix

Although oriented towards methods of solutions, this book also provides exact
expressions for the radiation field intensity and polarization for a number of standard
problems.
I am deeply grateful to many colleagues for their help and encouragement in
this undertaking. My sincere thanks go particularly to V. Bommier, V. V. Ivanov,
B. Rutily, and P. Zweifel, who have read significant parts of the book and have
taken the time to share with me their intimate knowledge of the field. I have adopted
many of their suggestions. My thanks go also to my family, especially to my mother
D. Piron-Lévy, for lasting encouragements; to my husband, Uriel, for reading the
full manuscript and playing the role of the non-specialist; and to L. Anusha, E. Lega,
and M. Sampoorna for their very generous help in the preparation of the figures. This
book would not exist without the continuous support of Springer teams in Europe
and India. I owe them a thousands thanks.

Nice Cedex 04, France Hélène Frisch

References

Ambartsumian, V.A.: Light scattering by planetary atmospheres. Astron. Zhurnal

19, 30–41 (1942)
Carleman, T.: Sur la résolution de certaines équations intégrales. Ark. Mat. Astr.
Fys. 16, 1–19 (1922)
Case, K.M.: Elementary solutions of the transport equation and their applications.
Ann. Phys. (New York) 9, 1–23 (1960)
Muskhelishvili, N.I.: Singular Integral Equations. Noordhoff, Groningen (Based on
the Second Russian Edition Published in 1946) (1953); Dover Publications
(1991)
Wiener, N., Hopf, D.: Über eine Klasse singulärer Integralgleichungen, Sitzungs-
berichte der Preussichen Akademie, Mathematisch-Physikalische Klasse 3. Dez.
1931, vol. 31, pp. 696–706 (ausgegeben 28. Januar 1932) (1931); English
translation: In: Paley, R.C., Wiener, N., Fourier transforms in the Complex
Domain. Am. Math. Soc. Coll. Publ., vol. XIX, pp. 49–58 (1934)
Contents

1 An Overview of the Content .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 1

1.1 Part I: Scalar Radiative Transfer Equations . . . . .. . . . . . . . . . . . . . . . . . . . 1
1.2 Part II: Scattering Polarization .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 4
1.3 Part III: Asymptotic Properties of Multiple Scattering . . . . . . . . . . . . . 6
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 8

Part I Scalar Radiative Transfer Equations

2 Radiative Transfer Equations . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 11
2.1 The Integro-Differential Radiative Transfer Equations .. . . . . . . . . . . . 12
2.1.1 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 17
2.1.2 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 18
2.1.3 The Diffuse Radiation Field. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 22
2.2 Integral Equations for the Source Function . . . . .. . . . . . . . . . . . . . . . . . . . 23
2.2.1 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 23
2.2.2 The Milne Problem . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 25
2.2.3 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 27
2.3 Neumann Series Expansion .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 30
2.4 The Green Function and Associated Functions .. . . . . . . . . . . . . . . . . . . . 32
2.4.1 Some Definitions . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 32
2.4.2 A Lemma on Wiener–Hopf Equations.. . . . . . . . . . . . . . . . . . . . 34
2.4.3 The Source Function and the Resolvent Function . . . . . . . . 35
2.4.4 Some Properties of the Green Function . . . . . . . . . . . . . . . . . . . 36
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 37
3 Exact Methods of Solution: A Brief Survey . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 41
3.1 The Infinite Medium and the Dispersion Function . . . . . . . . . . . . . . . . . 42
3.2 Exact Methods for a Semi-Infinite Medium . . . .. . . . . . . . . . . . . . . . . . . . 43
3.2.1 The Wiener–Hopf Method . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 43
3.2.2 Traditional Real-Space Methods . . . . . . .. . . . . . . . . . . . . . . . . . . . 44
3.2.3 The Singular Integral Equation Approach .. . . . . . . . . . . . . . . . 45
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 47

xi
xii Contents

4 Singular Integral Equations . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 51

4.1 The Inverse Laplace Transformation . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 52
4.1.1 The Half-Space Case . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 52
4.1.2 The Full-Space Case. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 54
4.2 The Direct Laplace Transformation . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 55
4.3 The Hilbert Transform Method of Solution .. . . .. . . . . . . . . . . . . . . . . . . . 57
4.3.1 A Special Case . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 58
4.3.2 The General Case . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 59
4.3.3 Application to Radiative Transfer.. . . . .. . . . . . . . . . . . . . . . . . . . 61
Appendix A: Hilbert Transforms .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 63
A.1 Definition and Properties. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 63
A.2 The Plemelj Formulae .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 64
A.3 The Plemelj Formulae for a Dirac Distribution .. . . . . . . . . . . . . . . . . . . . 66
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 67
5 The Scattering Kernel and Associated Auxiliary Functions . . . . . . . . . . . 69
5.1 The Kernel and Its Inverse Laplace Transform .. . . . . . . . . . . . . . . . . . . . 69
5.2 The Dispersion Function .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 73
5.2.1 Symmetries and Analyticity Properties .. . . . . . . . . . . . . . . . . . . 73
5.2.2 The Zeroes of the Dispersion Function .. . . . . . . . . . . . . . . . . . . 78
5.2.3 The Index κ . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 80
5.3 The Half-Space Auxiliary Function X(z) . . . . . . .. . . . . . . . . . . . . . . . . . . . 80
5.4 Some Properties of X(z) . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 83
5.4.1 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 83
5.4.2 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 84
5.4.3 Some Properties of the H -Function . . .. . . . . . . . . . . . . . . . . . . . 85
Appendix B: Properties of the Half-Space Auxiliary Function .. . . . . . . . . . . 88
B.1 Factorization Relation .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 89
B.2 Identities for the Boundary Values of X(z) . . . . .. . . . . . . . . . . . . . . . . . . . 89
B.3 Integral Equations for X(z) . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 91
B.4 Values of X(z) and H (z) . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 92
B.5 Moments of X(z) and H (z) . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 93
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 94
6 The Surface Green Function and the Resolvent Function . . . . . . . . . . . . . 97
6.1 Infinite Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 98
6.1.1 The Fourier Transform Method .. . . . . . .. . . . . . . . . . . . . . . . . . . . 98
6.1.2 The Inverse Laplace Transform Method .. . . . . . . . . . . . . . . . . . 101
6.2 Semi-infinite Medium. The Inverse Laplace
Transform Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 102
6.2.1 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 103
6.2.2 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 105
6.3 The H -Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 108
Appendix C: The Direct Laplace Transform Method . .. . . . . . . . . . . . . . . . . . . . 110
C.1 Complete Frequency Redistribution .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 111
C.2 Monochromatic Scattering .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 113
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 114
Contents xiii

7 The Emergent Intensity and the Source Function . .. . . . . . . . . . . . . . . . . . . . 115

7.1 The Emergent Intensity . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 115
7.2 The Source Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 118
7.2.1 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 119
7.2.2 Monochromatic Scattering. A Fourier Inversion .. . . . . . . . . 123
7.3 Some Standard Results . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 126
7.3.1 Uniform Primary Source Term .. . . . . . . .. . . . . . . . . . . . . . . . . . . . 126
7.3.2 Exponential Primary Source Term . . . . .. . . . . . . . . . . . . . . . . . . . 128
7.3.3 Diffuse Reflection . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 130
Appendix D: The Monochromatic Source Function . . . . . . . . . . . . . . . . . . . . . . . 134
D.1 The Inverse Laplace Transform Method . . . . . . . .. . . . . . . . . . . . . . . . . . . . 135
D.2 The Direct Laplace Transform Method . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 139
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 142
8 Spectral Line with Continuous Absorption . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 143
8.1 The Radiative Transfer Equation . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 144
8.2 Properties of the Auxiliary Functions . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 147
8.2.1 The Dispersion Function . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 147
8.2.2 The Half-Space Auxiliary Function . . .. . . . . . . . . . . . . . . . . . . . 149
8.3 Some Exact Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 149
8.3.1 The Resolvent Function . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 150
8.3.2 The Line Source Function .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 151
8.3.3 The Emergent Intensity.. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 154
Appendix E: Kernels for Spectral Lines with a Continuous
Absorption .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 155
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 158
9 Conservative Scattering: The Milne Problem . . . . . . .. . . . . . . . . . . . . . . . . . . . 159
9.1 The Monochromatic Milne Problem . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 160
9.1.1 The Auxiliary Functions . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 162
9.1.2 The Source Function and the Emergent Intensity .. . . . . . . . 163
9.1.3 The Hopf Function . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 165
9.2 The Milne Problem for a Spectral Line . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 167
Appendix F: The Conservative Auxiliary Functions. Complete
Frequency Redistribution . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 169
F.1 Asymptotic Behavior of the Kernel . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 170
F.2 The Dispersion Function and the Half-Space Auxiliary
Function Near the Origin . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 171
Appendix G: The Milne Problem with the Inverse Laplace
Transform Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 174
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 176
10 The Case Eigenfunction Expansion Method. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 179
10.1 The Eigenfunctions and Eigenvalues .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 180
10.1.1 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 180
10.1.2 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 182
xiv Contents

10.2 The Diffuse Reflection Problem .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 184

10.2.1 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 184
10.2.2 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 188
10.3 Further Applications of the Case Method .. . . . . .. . . . . . . . . . . . . . . . . . . . 190
10.3.1 A Complete Redistribution Full-Space Problem .. . . . . . . . . 190
10.3.2 A Half-Space Problem with Internal Sources . . . . . . . . . . . . . 192
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 194
√
11 The -law √and the Nonlinear H -Equation . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 195
11.1 The -law .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 195
11.1.1 A Quadratic Approach . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 197
11.1.2 A Green Function Approach . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 198
11.2 Construction of the Nonlinear H -Equation . . . . .. . . . . . . . . . . . . . . . . . . . 200
11.2.1 Scattering Coefficient Method . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 200
11.2.2 Exponential Primary Source Method . . . . . . . . . . . . . . . . . . . . . . 204
11.3 Some Properties of the H -Equation .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 206
11.3.1 A Factorization Relation . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 206
11.3.2 Uniqueness and Alternative Nonlinear H -Equations . . . . . 207
Appendix H: An Elementary Construction of the H -Function . . . . . . . . . . . . 211
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 213
12 The Wiener–Hopf Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 215
12.1 An Overview of the Wiener–Hopf Method . . . . .. . . . . . . . . . . . . . . . . . . . 216
12.2 Full Description of the Wiener–Hopf Method . .. . . . . . . . . . . . . . . . . . . . 220
12.2.1 Decomposition Formula .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 220
12.2.2 Analyticity of Complex Fourier Transforms . . . . . . . . . . . . . . 222
12.2.3 Properties of the Dispersion Function V (z) . . . . . . . . . . . . . . . 224
12.2.4 Factorization of the Dispersion Function .. . . . . . . . . . . . . . . . . 224
12.2.5 Decomposition of the Inhomogeneous Term .. . . . . . . . . . . . . 228
12.2.6 Fourier Transform of the Source Function . . . . . . . . . . . . . . . . 229
12.3 The Resolvent Function .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 231
12.4 The Milne Problem .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 233
12.5 The Wiener–Hopf Method for Spectral Lines . .. . . . . . . . . . . . . . . . . . . . 237
12.5.1 Factorization of V (z) . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 238
12.5.2 The Fourier Transform of the Source Function . . . . . . . . . . . 239
12.5.3 The Emergent Intensity.. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 240
12.6 The Sommerfeld Half-Plane Diffraction Problem . . . . . . . . . . . . . . . . . . 241
Appendix I: The Auxiliary Function Vu (z) . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 246
I.1 Explicit Expression of Vu (z). . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 247
I.2 The Auxiliary Functions Vu (z) and X(z) . . . . . . .. . . . . . . . . . . . . . . . . . . . 248
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 250

Part II Scattering Polarization

13 The Scattering of Polarized Radiation . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 255
13.1 Description of a Polarized Radiation .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 256
Contents xv

13.2 The Rayleigh Scattering Phase Matrix . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 261

13.2.1 The Scattering of a Collimated Radiation Beam .. . . . . . . . . 262
13.2.2 The Rayleigh Phase Matrix in a Fixed
Reference Frame .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 264
13.3 Resonance Scattering of Spectral Lines . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 271
13.3.1 Populations and Coherences . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 272
13.3.2 The Scattering Phase Matrix . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 274
13.4 The Hanle Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 277
13.4.1 Some General Properties . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 277
13.4.2 The Hanle Phase Matrix .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 279
Appendix J: Spectral Details of Resonance Scattering .. . . . . . . . . . . . . . . . . . . . 282
Appendix K: Spectral Details of the Hanle Effect.. . . . .. . . . . . . . . . . . . . . . . . . . 285
K.1 Magnetic Field Reference Frame . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 286
K.2 Atmospheric Reference Frame . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 289
K.3 Approximations for the Redistribution Matrix ... . . . . . . . . . . . . . . . . . . . 289
Appendix L: The Hanle Effect with the Classical Harmonic
Oscillator Model.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 290
L.1 The Harmonic Oscillator Model . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 290
L.2 Polarization Direction and Polarization Rate . . .. . . . . . . . . . . . . . . . . . . . 293
L.3 Frequency Redistribution . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 295
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 300
14 Polarized Radiative Transfer Equations . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 305
14.1 Rayleigh Scattering. The Radiative Transfer Equation .. . . . . . . . . . . . 307
14.2 Conservative Rayleigh Scattering . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 308
14.2.1 The Polarized Milne Problem .. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 309
14.2.2 The Diffuse Reflection Problem . . . . . . .. . . . . . . . . . . . . . . . . . . . 311
14.3 Rayleigh Scattering. A (KQ) Expansion of the Stokes
Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 312
14.3.1 Cylindrically Symmetric Radiation Field . . . . . . . . . . . . . . . . . 312
14.3.2 Azimuthally Dependent Radiation Field . . . . . . . . . . . . . . . . . . 316
14.4 Resonance Polarization of Spectral Lines . . . . . . .. . . . . . . . . . . . . . . . . . . . 321
14.4.1 One-Dimensional Medium .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 323
14.4.2 Multi-Dimensional Medium . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 327
14.5 The Hanle Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 330
14.5.1 The Hanle Redistribution Matrix . . . . . .. . . . . . . . . . . . . . . . . . . . 331
14.5.2 One-Dimensional Medium .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 334
14.5.3 Multi-Dimensional Medium . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 338
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 339
√
15 The -Law, the Nonlinear H-Equation, and Matrix
Singular Integral Equations . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 341
15.1 The Source
√ Vector: Its Derivative and Matrix Representation . . . . . 342
15.2 The -Law .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 344
15.2.1 Rayleigh Scattering and Resonance Polarization . . . . . . . . . 345
15.2.2 The Hanle Effect .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 347
xvi Contents

15.3 The Green Matrix and the Resolvent Matrix .. . .. . . . . . . . . . . . . . . . . . . . 349

15.3.1 Rayleigh Scattering and Resonance Polarization . . . . . . . . . 349
15.3.2 The Hanle Effect .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 351
15.4 Construction of the H-Equation . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 352
15.4.1 Rayleigh Scattering and Resonance Polarization . . . . . . . . . 353
15.4.2 The Hanle Effect .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 355
15.5 Alternative Definitions of the H-Matrix .. . . . . . . .. . . . . . . . . . . . . . . . . . . . 356
15.5.1 Exponential Primary Source . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 357
15.5.2 Uniform Primary Source . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 357
15.6 Singular Integral Equations .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 359
15.6.1 Rayleigh Scattering and Resonance Polarization . . . . . . . . . 360
15.6.2 The Hanle Effect .. . . . . .√ . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 361
15.7 Factorization Relations and the -Law .. . . . . . .. . . . . . . . . . . . . . . . . . . . 362
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 364
16 Conservative Rayleigh Scattering: Exact Solutions.. . . . . . . . . . . . . . . . . . . . 365
16.1 Radiative Transfer Equations . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 366
16.2 The Dispersion Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 369
16.3 Construction of the Auxiliary X-matrix . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 370
16.4 The Resolvent Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 373
16.5 The H-matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 377
16.6 The Polarized Milne Problem.. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 378
16.6.1 Some Physical Properties . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 379
16.6.2 The Source Vector . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 381
16.6.3 The Generalized Hopf Functions . . . . . .. . . . . . . . . . . . . . . . . . . . 382
16.6.4 The Emergent Radiation Field . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 385
16.6.5 The Stokes Parameters I and Q . . . . . . .. . . . . . . . . . . . . . . . . . . . 386
16.7 The Diffuse Reflection Problem .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 389
16.7.1 The Emergent Radiation Field . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 390
16.7.2 The Source Vector and the Radiation Field at Infinity .. . . 392
Appendix M: Properties of the Auxiliary Functions . . .. . . . . . . . . . . . . . . . . . . . 394
M.1 The Functions L1 (z) and L2 (z) . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 395
M.2 Some Properties of the Matrix L(ν) . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 397
M.3 The Auxiliary Functions X1 (z) and X2 (z) . . . . . .. . . . . . . . . . . . . . . . . . . . 398
M.3.1 Construction of X1 (z) and X2 (z) . . . . . .. . . . . . . . . . . . . . . . . . . . 398
M.3.2 Factorizations Relations . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 399
M.3.3 Nonlinear Integral Equations.. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 400
M.3.4 The Functions Hl (z) and Hr (z) . . . . . . . .. . . . . . . . . . . . . . . . . . . . 401
M.3.5 Moments of Hl (μ) and Hr (μ) . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 402
Appendix N: The Milne Integral Equation . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 404
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 409
17 Scattering Problems with No Exact Solution I: The Auxiliary
Matrices.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 411
17.1 Non-Conservative Rayleigh Scattering .. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 412
17.1.1 The Dispersion Matrix . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 413
17.1.2 The Dispersion Function . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 414
Contents xvii

17.1.3 The Auxiliary X-Matrix . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 417

17.1.4 A Factorization of the Dispersion Matrix . . . . . . . . . . . . . . . . . 419
17.2 Resonance Polarization . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 421
17.2.1 The Dispersion Matrix . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 422
17.2.2 The Auxiliary X-Matrix . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 423
17.3 The Hanle Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 424
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 426
18 Scattering Problems with No Exact Solution II: The
Resolvent Matrix, the H-Matrix, and the I-Matrix... . . . . . . . . . . . . . . . . . . . 427
18.1 Non-Conservative Rayleigh Scattering .. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 428
18.1.1 The Resolvent Matrix . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 428
18.1.2 The H-Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 432
18.1.3 The H-Equation in the Complex Plane . . . . . . . . . . . . . . . . . . . . 434
18.1.4 Uniqueness of the Solution of the H-Equation .. . . . . . . . . . . 436
18.1.5 An Alternative H-Equation .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 437
18.1.6 The Emergent Radiation Field and the I-Matrix . . . . . . . . . . 439
18.2 Resonance Polarization . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 442
18.2.1 The Resolvent Matrix . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 443
18.2.2 The H-Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 444
18.2.3 Nonlinear Integral Equations.. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 445
18.3 The Hanle Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 447
18.3.1 The H-Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 448
18.3.2 Nonlinear Integral Equations.. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 449
Appendix O: The Singular Integral Equation for G̃(p) . . . . . . . . . . . . . . . . . . . . 451
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 455

Part III Asymptotic Properties of Multiple Scattering

19 Asymptotic Properties of the Scattering Kernel K(τ ) .. . . . . . . . . . . . . . . . . 459
19.1 The Monochromatic Kernel . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 461
19.2 The Complete Frequency Redistribution Kernel.. . . . . . . . . . . . . . . . . . . 461
19.2.1 Asymptotic Behavior at Large Optical Depths .. . . . . . . . . . . 462
19.2.2 The Inverse Laplace Transform Near the Origin .. . . . . . . . . 463
19.2.3 The Fourier Transform Near the Origin . . . . . . . . . . . . . . . . . . . 464
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 466
20 Large Scale Radiative Transfer Equations . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 467
20.1 Asymptotic Analysis of the Source Function . . .. . . . . . . . . . . . . . . . . . . . 468
20.1.1 Monochromatic Scattering . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 470
20.1.2 Complete Frequency Redistribution . . .. . . . . . . . . . . . . . . . . . . . 470
20.1.3 Higher Order Terms . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 472
20.2 Asymptotic Analysis in Fourier Space . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 473
20.3 The Thermalization Length .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 474
20.4 Ordinary and Anomalous Diffusion .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 477
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 478
xviii Contents

21 The Photon Random Walk . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 479

21.1 The Mean Displacement After n Steps . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 481
21.1.1 Mean Displacement: Method I . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 481
21.1.2 Mean Displacement: Method II . . . . . . . .. . . . . . . . . . . . . . . . . . . . 486
21.1.3 Application to Normal Diffusion and Lévy Walks . . . . . . . . 487
21.2 The Mean Positive Maximum After n Steps . . . .. . . . . . . . . . . . . . . . . . . . 491
21.2.1 A Wiener–Hopf Integral Equation .. . . .. . . . . . . . . . . . . . . . . . . . 491
21.2.2 A Precise Asymptotics √ for the Mean Maximum . . . . . . . . . . 494
21.3 A Probabilistic Proof of the -Law.. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 499
Appendix P: Universality of Escape from a Half-Space . . . . . . . . . . . . . . . . . . . 502
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 505
22 Asymptotic Behavior of the Resolvent Function . . . .. . . . . . . . . . . . . . . . . . . . 507
22.1 The Infinite Medium Resolvent Function . . . . . . .. . . . . . . . . . . . . . . . . . . . 507
22.1.1 Infinite Medium with a Plane Source ... . . . . . . . . . . . . . . . . . . . 508
22.1.2 Infinite Medium with a Point Source . .. . . . . . . . . . . . . . . . . . . . 511
22.2 Semi-Infinite Medium .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 513
Reference .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 514
23 The Asymptotics of the Diffusion Approximation . .. . . . . . . . . . . . . . . . . . . . 515
23.1 The Rescaled Equation .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 516
23.2 The Interior Radiation Field . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 517
23.3 An Improved Diffusion Equation .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 520
23.4 The Boundary Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 522
23.4.1 Boundary Conditions for the Interior Solution .. . . . . . . . . . . 524
23.4.2 The Emergent Intensity.. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 525
23.5 The Eddington Approximations . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 526
23.6 Determination of the Critical Size . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 529
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 531
24 The Diffusion Approximation for Rayleigh Scattering .. . . . . . . . . . . . . . . . 533
24.1 The Radiative Transfer Equation . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 534
24.2 The Interior Radiation Field Expansion .. . . . . . . .. . . . . . . . . . . . . . . . . . . . 536
24.3 The Boundary Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 539
24.3.1 The Boundary Layer as a Diffuse Reflection Problem.. . . 539
24.3.2 Matching of the Interior and Boundary Layer Fields . . . . . 541
24.4 The Emergent Radiation Field . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 543
24.4.1 The Polarization Rate. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 544
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 548
25 Anomalous Diffusion for Spectral Lines . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 549
25.1 Interior and Boundary Layer Expansions . . . . . . .. . . . . . . . . . . . . . . . . . . . 550
25.2 The Radiation Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 552
25.3 Resonance Polarization . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 553
25.4 Scaling Laws for a Slab .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 556
25.5 Mean Number of Scatterings and Mean Path Length . . . . . . . . . . . . . . 558
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 562
Contents xix

26 Asymptotic Results for Partial Frequency Redistribution . . . . . . . . . . . . . 563

26.1 RI Asymptotic Behavior . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 565
26.2 RIII Asymptotic Behavior .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 568
26.3 RV Asymptotic Behavior . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 569
26.4 RII Asymptotic Behavior.. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 572
26.4.1 Diffusion Equation for the Source Function . . . . . . . . . . . . . . 574
26.4.2 Diffusion Equation for the Radiation Field.. . . . . . . . . . . . . . . 576
26.4.3 RII Scaling Laws .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 578
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 580

Author Index.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 583

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 587
List of Figures

Fig. 2.1 Left panel: the space variable τ and the inclination angle
θ of a ray with direction n. The axis z is normal to the
surface of the medium. Right panel: the heliocentric
interpretation of the angle θ . The Sun is viewed from the
right. Observations at disk center correspond to θ = 0 and
those at the limb to θ = π/2 . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 14
Fig. 2.2 The Doppler profile, and several Voigt profiles with
different values of the parameter a. The left panel shows
ϕ(x) and the right panel log ϕ(x). In the right panel, the
Lorentzian wing regime characterized by ϕ(x) ∼ 1/x 2
can be observed for |x| larger than (− ln a)1/2 .. . . . . . . . . . . . . . . . . . . . 21
Fig. 2.3 Complete frequency redistribution. The integration
domains (ξ, x) and (x, ξ ), with ξ = μ/ϕ(x) . . .. . . . . . . . . . . . . . . . . . . . 29
Fig. A.1 Integration contour for the Plemelj formulae. The
contribution from the singular point is obtained by letting
ρ → 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 65
Fig. 5.1 The function g(ξ ) for the Doppler profile and the Lorentz
profile, 1/(π(1 + x 2 )), in linear scales in the upper panel
and in log-log scales in the lower panel. The value of
g(ξ ) is constant for ξ ≤ 1/ϕ(0). The algebraic behavior
of g(ξ ) for ξ → ∞ given in Eqs. (5.8) and (5.9) can be
observed in the lower panel. The Voigt profile has the
same algebraic behavior as the Lorentz profile .. . . . . . . . . . . . . . . . . . . . 71
Fig. 5.2 The function k(ν) for the Doppler profile and the Lorentz
profile, 1/(π(1 + x 2 )), in linear scales in the upper panel
and in log-log scales in the lower panel. The angular
point is located at ν = ϕ(0). For the Doppler and Lorentz
profiles, k(ν) tends to zero as 1/ν for ν → ∞. For
ν → 0, one can observe in the lower panel the algebraic
behaviors given in Eqs. (5.8) and (5.9). The Voigt profile
has the same algebraic behaviors as the Lorentz profile . . . . . . . . . . . 72

xxi
xxii List of Figures

Fig. 5.3 Branch cuts of the monochromatic dispersion function.

The solid black lines on the real axis show the branch cuts
of L(z) . For complete frequency redistribution, L(z) is
singular along the full real axis, except at the origin . . . . . . . . . . . . . . . 74
Fig. 5.4 Monochromatic scattering. The function λ(ν), real
part of L± (ν), defined in Eq. (5.28), is shown here for
 = 1/8 = 0.125. It becomes infinite for ν = ±1, takes
the value  at ν = 0 and has two zeroes in the interval
[−1, +1] at ±ν0 . They coalesce into a double zero at the
origin for  = 0. The zeroes at ±ν0 correspond to the two
zeroes of L(z) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 76
Fig. 5.5 Monochromatic scattering. Phase diagrams of L+ (ν) and
of its complex conjugate L− (ν) for ν between 1 and
+∞. The change of sign of [L+ (ν)] corresponds to the
change of sign of λ(ν) in the interval ν ∈ [1, ∞[. The
angle θ is the argument of L+ (ν). It varies between −π
and 0 for ν ∈ [1, ∞[. The destruction probability  has
the value 0.125 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 77
Fig. 5.6 Complete frequency redistribution. Sketch of the phase
diagrams of L+ (ν) and L− (ν) for a Lorentz profile and
 = 0. The phase diagram lies entirely to the right of the
imaginary axis since [L+ (ν)] = λ(ν) has no zero . . . . . . . . . . . . . . . . 77
Fig. 5.7 Contour for the determination of the number of zeroes
of the dispersion function L(z). For monochromatic
scattering, the end point of the cuts are at ν = ±1. For
complete frequency redistribution, they are at ν = 0 . . . . . . . . . . . . . . . 78
Fig. 5.8 The monochromatic function H (μ) for isotropic
scattering, with different √ values of . For
 = 0, H (μ) √ = 3μ as μ → ∞. For  = 0,
H (μ)
√ = 1/  as μ → ∞. For the chosen values of ,
1/  = 6.3, 3.2, 2.2, 1.6. The numerical data are from
Chandrasekhar (1960, p. 125) . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 87
Fig. B.1 Complete frequency redistribution. Closed contour for the
construction of the nonlinear integral equation satisfied
by X(z) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 91
Fig. 6.1 Contours for the determination of the infinite medium
Green function by application of the Fourier inversion
formula. Left panel: complete frequency redistribution;
right panel: monochromatic scattering . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 99
Fig. 6.2 Contours for the calculation of G̃(p), the Laplace
transform of the surface Green function G(τ ). Left
panel: complete frequency redistribution. Right panel:
monochromatic scattering . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 108
List of Figures xxiii

Fig. C.1 Integration contours for the application of the Fourier

inversion formula. Left panel: calculation of the resolvent
function (τ ) for complete frequency redistribution.
Right panel: calculation of the source function S(τ ) for
monochromatic scattering . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 112
Fig. 7.1 The source function for a uniform primary source Q∗ = 
and several choices of . The absorption profile is a Voigt
−2
profile with parameter √ a = 10 . The surface value S(0)
follows the exact -law . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 127
Fig. 7.2 The source function for an exponential primary source
Q∗ (τ ) = exp(−10−4 τ ) and  = 10−6 , for a Voigt profile
and a Doppler profile. For the latter, the straight line
shows the asymptotic behavior of the boundary layer
solution (see also Fig. 9.1) . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 129
Fig. E.1 The function kβ (ν, β) for the Doppler profile and
different values of the parameter β, ratio of the continuous
absorption coefficient to the line absorption coefficient.
The function kβ (ν, β) is zero up to ν = β. It is defined in
Eq. (8.17) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 157
Fig. 9.1 Large τ asymptotic behavior of the source function for
conservative scattering. The function S(τ ) is calculated
with a uniform primary source  = 10−6 for a Doppler
profile and a Voigt profile. The straight lines show the
corresponding asymptotic behaviors, τ 1/2 (ln τ )1/4 and
τ 1/4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 169
Fig. F.1 Complete frequency redistribution. Sketch of the phase
diagrams of L+ (ν) and L− (ν) for  = 0. At the point
ν = 0, L± (0) = L(0) = 0 and the phase diagrams are
tangent to the vertical axis in the case of the Doppler
profile and and make an π/4 angle with this axis for a
Voigt profile. For ν → ∞, L± (∞) = L(∞) = 1 and the
phase diagrams are tangent to the vertical axis .. . . . . . . . . . . . . . . . . . . . 173
Fig. 12.1 Integration contour for the decomposition formula. The
vertical lines with an arrow-head indicate that f + (z) is
analytic in an upper half-plane and that f − (z) is analytic
in a lower half-plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 221
Fig. 12.2 The analyticity domains of V (z), Su (z), the Fourier
transform of S + (τ ), and of Sl (z), the Fourier transform
of S − (τ ). The dashed line a has been placed arbitrarily
below the line +1. The vertical lines with two arrows
indicate a strip of analyticity, those with a simple arrow
indicate a half-plane analyticity. The functions V (z)
and Su (z) have a common strip of analyticity defined
by a < (z) < 1. The function V (z) is analytic in
−1 < (z), +1 and has two zeroes at ±i ν0 . . . .. . . . . . . . . . . . . . . . . . . . 223
xxiv List of Figures

Fig. 12.3 The figure shows the strips of analyticity of V (z), the
dispersion function, and of V ∗ (z) defined in Eq. (12.34).
It shows also the analyticity half-planes of Vu (z) and
Vl (z), which satisfy V (z) = Vu (z)/Vl (z), and of Vu∗ (z)
and Vl∗ (z), which satisfy V ∗ (z) = Vu∗ (z)/Vl∗ (z) . . . . . . . . . . . . . . . . . . 225
Fig. 12.4 Analyticity half-planes of the inhomogeneous term
Q̂∗ (z) and of the functions Gu (z) and Gl (z) defined in
Eq. (12.47). The thick solid line indicates the branch cut
of Vl (z) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 229
Fig. 12.5 The analyticity half-planes of the functions in the left
and right hand-sides of Eq. (12.16) and their common
analyticity strip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 230
Fig. 12.6 The Wiener–Hopf method for the Milne problem. The
half-plane analyticity domains of the functions in the left
and right hand-sides of Eq. (12.73) and their common
strip of analyticity. The function Vu (z) has a double zero
at the origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 235
Fig. 12.7 The Sommerfeld diffraction problem: (a) the geometry;
(b) the boundary conditions for the field ϕ(x, y) and
for ϕ (x, 0± ) = ∂ϕ(x, 0± )/∂y, the normal derivative of
ϕ(x, y) on the axis y = 0. The notation 0± means that
y → 0 by positive or negative values. The continuity
across the axis y = 0 is indicated by a vertical line with
two arrows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 241
Fig. 12.8 The Sommerfeld diffraction problem: the factorization of
γ = (k − λ)1/2 (k + λ)1/2 . The principal determination of
(k + λ)1/2 is analytic in (k) > 0 and that of (k − λ)1/2
in (k) < 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 245
Fig. I.1 Monochromatic scattering. (a) Contours for obtaining an
explicit expression of Vu (z) in terms of the logarithm
of the dispersion function. The point z lies in the upper
half-plane (z) > δ > −ν0 . (b) Contour for obtaining the
relation between Vu (z) and X(z). The point z lies in the
upper half-plane (z) > −1 . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 247
Fig. 13.1 Ellipse described by the tip of the electric field vector in
the plane orthogonal to the direction of propagation. The
angle  (0 ≤  ≤ π) specifies the orientation of the
ellipse and the angle χ (−π/4 ≤ χ ≤ π/4) the ellipticity
and the sense in which the ellipse is being described . . . . . . . . . . . . . . 257
Fig. 13.2 Rotation of the reference direction by an angle α . . . . . . . . . . . . . . . . . . 259
Fig. 13.3 Geometry of a single scattering. The components Ir
and Il are respectively perpendicular and parallel to the
scattering plane (,  ) . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 262
List of Figures xxv

Fig. 13.4 The atmospheric reference frame. The axis Oz is along

the direction normal to the surface. A direction  is
defined by its colatitude θ and azimuth χ. In the plane
perpendicular to , ea is the reference direction and eb is
perpendicular to ea . γ is the angle between the tangent to
the meridian plane and ea . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 264
Fig. 13.5 Sketch of the Zeeman displacement of the magnetic
sublevels for a normal Zeeman triplet. The horizontal axis
is the magnetic field intensity B. The Hanle effect can be
observed as long as the Zeeman sublevels are overlapping
each other. The magnetic field intensity Btyp corresponds
to a Hanle factor H of order unity . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 277
Fig. L.1 The spherical unit vectors e0 , e+1 , and e+1 . The vector e0
is aligned with the magnetic field vector B . . . . .. . . . . . . . . . . . . . . . . . . . 292
Fig. L.2 Motion of the damped classical oscillator in the plane
perpendicular to the direction of the magnetic field.
The parametric plot shows the time evolution of the
component xy (t) versus the component xx (t), defined in
Eq. (L.14). The left panel shows a rosette with a Hanle
parameter H = 1 and the right one a daisy with H = 9.
The curves have been calculated with γ = 1, 2πν0 = 50,
and a maximum value t = 2.5 for the rosette and t = 1.5
for the daisy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 294
Fig. 14.1 The elements of the kernel matrix K(τ ) for the Doppler
profile, with the depolarization parameter set equal to 1.
The left panel shows the elements Kij (τ ) and the right
panel shows log K11 (τ ) (left vertical axis) and the ratios
K12 (τ )/K11 (τ ) and K22 (τ )/K11 (τ ) (right vertical axis) . . . . . . . . . . . . 327
Fig. 16.1 The functions Hl (μ) and Hr (μ) for Rayleigh scattering,
and the function H (μ) for conservative monochromatic
scattering.
√ For μ → ∞, √ Hl (μ) increases at infinity
as 5μ and H (μ) as 3μ, while Hr (μ) tends to the
constant value 0.1414. The linear growth of Hl (μ) and
H (μ) is already installed for μ around one. The data are
from Chandrasekhar (1960, p. 125 and p. 248) .. . . . . . . . . . . . . . . . . . . . 373
Fig. 20.1 The thermalization length for monochromatic scattering
and complete frequency redistribution. The figure shows
the source function S(τ ) for a uniform primary source 
with  = 10−6 . The thermalization length, distance from
the surface at which the source√ function S(τ ) reaches its
saturation value, is of order 1/  for monochromatic
scattering, 1/ for the Doppler profile, and a/( 2) for the
Voigt profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 475
xxvi List of Figures

Fig. 20.2 Ordinary diffusion and anomalous diffusion. The left

panel, shows a diffusive-type random walk. All the
random steps have the same constant value. The right
panel shows an anomalous random walk. The random
steps are calculated with a probability distribution having
the form 1/(1 + |x|β+1 ), with β = 1, mimicking the
Doppler frequency redistribution . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 476
Fig. 21.1 A typical discrete random walk on a line. The vertical
axis represents the position on the line of a random walker
starting at 0. After n steps, the random walker has the
position xn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 480
Fig. 21.2 Examples of Lévy walks. The probability distribution has
the form 1/(1 + |x|β+1). In the left panel, β = 2 and in
the right panel β = 1.5, which is a model for a Holtsmark
distribution. The case β = 1 is shown in the right panel
of Fig. 20.2. The particles are starting at the origin (0,0).
Each graphic shows 104 steps . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 490
Fig. 21.3 A typical random walk on a line, for a random walker
starting at the point x. Mn indicates the maximum
position reached while performing n steps. The first step
brings the random walker from x to x1 , x1 ≤ y . . . . . . . . . . . . . . . . . . . . 492
Fig. 22.1 Geometry for establishing the relation between of the
point source and the plane source resolvent functions . . . . . . . . . . . . . 512
Fig. 23.1 At each point r b of the surface, the boundary layer is
a plane-parallel, semi-infinite medium extending from
s = 0 to s = −∞. The surface of the boundary layer is
tangential to the surface of the medium at r b . . .. . . . . . . . . . . . . . . . . . . . 520
Fig. 24.1 The surface polarization rate (in percent). A comparison
between the asymptotic predictions of Eq. (24.60) (the
dashed lines) and numerical solutions of the polarized
transfer equation, for different values of  (the solid
lines). The √ model is a slab with an optical thickness
larger than , an unpolarized and uniform primary
source and no incident radiation. From top to bottom,
 = 10−5 , 10−3 , 10−2 , 10−1 . The asymptotic result is
not plotted for  = 10−1. The Chandrasekhar limit is
graphically indistinguishable from the  = 10−5 results . . . . . . . . . . . 546
Fig. 26.1 The RI partial redistribution function. This figure show
the ratio r̄I (x, x )/ϕ(x) as a function of x for different
values of x; x and x are respectively the incident and
scattered dimensionless frequencies . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 566
List of Figures xxvii

Fig. 26.2 The RV partial redistribution function. This figure shows

the partial redistribution function rV (ξ, ξ ) in the atomic
rest frame given in Eq. (26.22) for the damping parameter
γ1 = 0.05, the incident frequency ξ = 2, and three
different values of the ratio γ2 /γ1 . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 569
Fig. 26.3 The partial RII redistribution function. This figure shows
the ratio r̄II (x, x )/ϕ(x) as a function of x for different
values of the frequency x; x and x are respectively the
incident and scattered dimensionless frequencies . . . . . . . . . . . . . . . . . . 573
List of Tables

Table 5.1 Monochromatic scattering and complete frequency

redistribution. Real part, imaginary part, and argument
of L+ (ν) for  = 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 76
Table B.1 The functions X(z), X∗ (z) = (ν0 − z)X(z), and H (z)
for  = 0: values at z = 0 and behavior at infinity. For
monochromatic scattering, ν0 ∈ [0, 1] is the positive
√ dispersion function L(z). For small values of
root of the
, ν0 3. The behavior at infinity of H (z) for  = 0
is discussed in Chap. 9. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 93
Table 6.1 Infinite medium. G∞ (τ ) is the infinite medium Green
function and ∞ (τ ) the resolvent function. Columns
(2) and (3) correspond to the Fourier transform and
inverse Laplace transform of the functions in column
(1). Column (4) gives the primary source term for the
convolution integral equation satisfied by the functions
in column (1). Columns (5) and (6) give the Fourier and
inverse Laplace transforms of the functions in column
(4). The notation—stands for undefined functions . . . . . . . . . . . . . . . 98
Table 6.2 Semi-infinite medium. G(τ ) is the surface Green
function, and (τ ) the resolvent function. Columns (2)
and (3) give the direct and inverse Laplace transforms
of the functions in column (1). Column (4) gives
the primary source term for the convolution integral
equation satisfied by the functions in column (1).
Columns (5) and (6) give the direct and inverse
Laplace transforms of the functions in column (4). The
notation—stands for nonexistent functions . . .. . . . . . . . . . . . . . . . . . . . 103

xxix
xxx List of Tables

Table 7.1 Semi-infinite medium. S(τ ) denotes the source function,

Q∗ (τ ) the primary source term and K(τ ) the kernel of
the Wiener–Hopf integral equation for S(τ ). Columns
(2) and (3) correspond to the direct and inverse Laplace
transforms of the functions in column (1). Column (4)
lists examples of primary source terms and columns (5)
and (6) their direct and inverse Laplace transforms . . . . . . . . . . . . . . . 119
Table 8.1 The line source function at the surface and at infinity.
The effects of a continuous absorption and of a
continuous emission. β is the ratio of the continuous to
line absorption coefficient, ¯ the effective destruction
probability, and ρB the continuous emission. The
parameters β, , ρ and B are constants . . . . . . . .. . . . . . . . . . . . . . . . . . . . 153
Table F.1 Complete frequency redistribution. Real part, imaginary
part, and argument of L+ (ν) for  = 0 . . . . . . . .. . . . . . . . . . . . . . . . . . . . 172
Table 15.1 Notation for inverse Laplace transforms of vectors and
matrices depending on the optical depth τ . . . .. . . . . . . . . . . . . . . . . . . . 359
Table 16.1 The Rayleigh Milne problem. The surface value and
behavior at infinity of the two components of the source
vector for different representation of the polarized
radiation field. Sl and Sr correspond to the (Il , Ir )
representation and SI and SQ correspond to the (I, Q)
one. For comparison the first line shows the source
function S(τ ) for the scalar Milne problem. For the
scalar √case, s1 = 3F /(4π) and for Rayleigh scattering,
s1R = 6F /(8π) with F the total flux, defined for
Rayleigh scattering in Eq. (16.77). We recall that
c = β/α = 0.873 and q = 2/α = 0.690 . . . . . .. . . . . . . . . . . . . . . . . . . . 388
Table M.1 Real parts, imaginary parts, and arguments of L+ 1 (ν)
and L+ 2 (ν) along the cut [1, ∞[ . . . . . . . . . . . . . . .. . .................. 396
Table M.2 The functions L1,2 (z), X1,2 (z) and Hl,r (z) at
three special points z = 0, 1, ∞. We recall that
zX1 (−1/z) = Hl (z) and X2 (−1/z) = Hr (z). The
values of Hl,r (1) are from Chandrasekhar. The values of
X1,2 (1) are derived from the values of Hl,r (1) and the
factorization relations in Eq. (M.24) . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 399
Table M.3 Moments of Hl (μ) and Hr (μ) and some numerical
constants. The two first moments of Hl (μ) and Hr (μ),
the constants q and c, and the values of Hl (1) and Hr (1)
are from (Chandrasekhar 1960, p. 248), rounded to 3
significant digits. This reference provides also tables for
Hl (μ) and Hr (μ), μ ∈ [0, 1], which we have used to
calculate numerically the second and third-moments . . . . . . . . . . . . . 402
List of Tables xxxi

Table 18.1 The matrices G̃(z), H(z), G̃R (z), and HR (z) at the origin
and at infinity. The matrix R = GR (∞) is a rotation
matrix, which can be calculated by solving numerically
the nonlinear integral equations for G̃R (z) or HR (z). The
matrix E = diag[, Q ] is equal to L(0). The I-matrix is
defined by I(0, z) = HR (z) . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 439
Table 19.1 The asymptotic behaviors of K(τ ) for τ → ∞, g(ξ ) for
ξ → ∞, k(ν) for ν → 0, and K̂(k) for k → 0. The
kernel is an even function of τ . Here τ , ξ , and ν are
positive. For monochromatic scattering, k(ν) is zero in
the interval [0, 1[ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 465
Table 19.2 The index α for the Doppler and Voigt profiles and the
asymptotic expressions of f (y) and f (y) for y large.
The function f (y) is the inverse function of the profile
ϕ(x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 466
Table 22.1 Infinite medium. Behavior at large optical depths of the
resolvent function for small values of , the destruction
probability per scattering. The constant a is the Voigt
parameter of the line. The nearly conservative zone
corresponds to 1 τ < τeff and the strong absorption
zone to τ > τeff . For monochromatic scattering τeff ∼ ν0 . . . . . . . . 509
Table 22.2 Semi-infinite medium. Asymptotic behavior of the
resolvent function in the nearly conservative and strong
absorption zones. The thermalization lengths τeff are
given in Table 22.1. For monochromatic scattering
τeff ∼ ν0 . For  = 0, the monochromatic √ resolvent
function has exactly the value 3 at infinity . .. . . . . . . . . . . . . . . . . . . . 513
Table 24.1 Moments of Hl (μ) and Hr (μ) and other numerical
constants. The two first moments of Hl (μ) and Hr (μ),
the constants q and c, and the values of Hl (1) and Hr (1)
are from Chandrasekhar (1960, p. 248). This reference
provides also tables for Hl (μ) and Hr (μ), μ ∈ [0, 1],
used to calculate numerically the third and fourth order
moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 544
Table 25.1 Scaling laws for complete frequency redistribution
with a Doppler profile. In the first column, the small
expansion parameter; in row 1: β = 0 and T = ∞, in
row 2:  = 0 and T = ∞, and in row 3: β =  = 0. The
four subsequent columns show τeff , the characteristic
scale of variation of the radiation field, xc , the
characteristic frequency defined by τeff ϕ(xc ) ∼ 1, the
mean number of scatterings N, and the mean path
length L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 559
xxxii List of Tables

Table 25.2 Scaling laws for complete frequency redistribution with

a Voigt profile. a Voigt parameter of the line. Same
presentation as in Table 25.1 . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 559
Table 26.1 Scaling laws for the RII partial frequency redistribution.
a parameter of the line profile. The expansion parameter
is  for an infinite medium and 1/T for a slab with total
optical thickness T . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 579